-
1.
Associations of obesity phenotypes with weight change, cardiometabolic benefits, and type 2 diabetes incidence during a lifestyle intervention: results from the PREVIEW study.
Zhu, R, Huttunen-Lenz, M, Stratton, G, Handjieva-Darlenska, T, Handjiev, S, Sundvall, J, Silvestre, MP, Jalo, E, Pietiläinen, KH, Adam, TC, et al
International journal of obesity (2005). 2023;(9):833-840
Abstract
BACKGROUND/OBJECTIVES Some individuals with overweight/obesity may be relatively metabolically healthy (MHO) and have a lower risk of cardiovascular disease than those with metabolically unhealthy overweight/obesity (MUO). We aimed to compare changes in body weight and cardiometabolic risk factors and type 2 diabetes incidence during a lifestyle intervention between individuals with MHO vs MUO. METHODS This post-hoc analysis included 1012 participants with MHO and 1153 participants with MUO at baseline in the randomized trial PREVIEW. Participants underwent an eight-week low-energy diet phase followed by a 148-week lifestyle-based weight-maintenance intervention. Adjusted linear mixed models and Cox proportional hazards regression models were used. RESULTS There were no statistically significant differences in weight loss (%) between participants with MHO vs MUO over 156 weeks. At the end of the study, weight loss was 2.7% (95% CI, 1.7%-3.6%) in participants with MHO and 3.0% (2.1%-4.0%) in those with MUO. After the low-energy diet phase, participants with MHO had smaller decreases in triglyceride (mean difference between MHO vs MUO 0.08 mmol·L-1 [95% CI, 0.04-0.12]; P < 0.001) but similar reductions in fasting glucose and HOMA-IR than those with MUO. However, at the end of weight maintenance, those with MHO had greater reductions in triglyceride (mean difference -0.08 mmol·L-1 [-0.12--0.04]; P < 0.001), fasting glucose, 2-hour glucose (difference -0.28 mmol·L-1 [-0.41--0.16]; P < 0.001), and HOMA-IR than those with MUO. Participants with MHO had smaller decreases in diastolic blood pressure and HbA1c and greater decreases in HDL cholesterol after weight loss than those with MUO, whereas the statistically significant differences disappeared at the end of weight maintenance. Participants with MHO had lower 3-year type 2 diabetes incidence than those with MUO (adjusted hazard ratio 0.37 [0.20-0.66]; P < 0.001). CONCLUSIONS Individuals with MUO had greater improvements in some cardiometabolic risk factors during the low-energy diet phase, but had smaller improvements during long-term lifestyle intervention than those with MHO.
-
2.
Longitudinal Associations of Dietary Sugars and Glycaemic Index with Indices of Glucose Metabolism and Body Fatness during 3-Year Weight Loss Maintenance: A PREVIEW Sub-Study.
Della Corte, K, Jalo, E, Kaartinen, NE, Simpson, L, Taylor, MA, Muirhead, R, Raben, A, Macdonald, IA, Fogelholm, M, Brand-Miller, J
Nutrients. 2023;(9)
Abstract
BACKGROUND Dietary sugars are often linked to the development of overweight and type 2 diabetes (T2D) but inconsistencies remain. OBJECTIVE We investigated associations of added, free, and total sugars, and glycaemic index (GI) with indices of glucose metabolism (IGM) and indices of body fatness (IBF) during a 3-year weight loss maintenance intervention. DESIGN The PREVIEW (PREVention of diabetes through lifestyle Intervention and population studies in Europe and around the World) study was a randomised controlled trial designed to test the effects of four diet and physical activity interventions, after an 8-week weight-loss period, on the incidence of T2D. This secondary observational analysis included pooled data assessed at baseline (8), 26, 52, 104 and 156 weeks from 514 participants with overweight/obesity (age 25-70 year; BMI ≥ 25 kg⋅m-2) and with/without prediabetes in centres that provided data on added sugars (Sydney and Helsinki) or free sugars (Nottingham). Linear mixed models with repeated measures were applied for IBF (total body fat, BMI, waist circumference) and for IGM (fasting insulin, HbA1c, fasting glucose, C-peptide). Model A was adjusted for age and intervention centre and Model B additionally adjusted for energy, protein, fibre, and saturated fat. RESULTS Total sugars were inversely associated with fasting insulin and C-peptide in all centres, and free sugars were inversely associated with fasting glucose and HbA1c (Model B: all p < 0.05). Positive associations were observed between GI and IGM (Model B: fasting insulin, HbA1c, and C-peptide: (all p < 0.01), but not for added sugars. Added sugar was positively associated with body fat percentage and BMI, and GI was associated with waist circumference (Model B: all p < 0.01), while free sugars showed no associations (Model B: p > 0.05). CONCLUSIONS Our findings suggest that added sugars and GI were independently associated with 3-y weight regain, but only GI was associated with 3-y changes in glucose metabolism in individuals at high risk of T2D.
-
3.
Socio-economic factors, mood, primary care utilization, and quality of life as predictors of intervention cessation and chronic stress in a type 2 diabetes prevention intervention (PREVIEW Study).
Huttunen-Lenz, M, Raben, A, Adam, T, Macdonald, I, Taylor, MA, Stratton, G, Mackintosh, K, Martinez, JA, Handjieva-Darlenska, T, Bogdanov, GA, et al
BMC public health. 2023;(1):1666
Abstract
BACKGROUND Sedentary lifestyle and unhealthy diet combined with overweight are risk factors for type 2 diabetes (T2D). Lifestyle interventions with weight-loss are effective in T2D-prevention, but unsuccessful completion and chronic stress may hinder efficacy. Determinants of chronic stress and premature cessation at the start of the 3-year PREVIEW study were examined. METHODS Baseline Quality of Life (QoL), social support, primary care utilization, and mood were examined as predictors of intervention cessation and chronic stress for participants aged 25 to 70 with prediabetes (n = 2,220). Moderating effects of sex and socio-economic status (SES) and independence of predictor variables of BMI were tested. RESULTS Participants with children, women, and higher SES quitted intervention earlier than those without children, lower SES, and men. Lower QoL, lack of family support, and primary care utilization were associated with cessation. Lower QoL and higher mood disturbances were associated with chronic stress. Predictor variables were independent (p ≤ .001) from BMI, but moderated by sex and SES. CONCLUSIONS Policy-based strategy in public health should consider how preventive interventions may better accommodate different individual states and life situations, which could influence intervention completion. Intervention designs should enable in-built flexibility in delivery enabling response to individual needs. TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT01777893.
-
4.
Gut microbiota predicts body fat change following a low-energy diet: a PREVIEW intervention study.
Jian, C, Silvestre, MP, Middleton, D, Korpela, K, Jalo, E, Broderick, D, de Vos, WM, Fogelholm, M, Taylor, MW, Raben, A, et al
Genome medicine. 2022;(1):54
Abstract
BACKGROUND Low-energy diets (LEDs) comprise commercially formulated food products that provide between 800 and 1200 kcal/day (3.3-5 MJ/day) to aid body weight loss. Recent small-scale studies suggest that LEDs are associated with marked changes in the gut microbiota that may modify the effect of the LED on host metabolism and weight loss. We investigated how the gut microbiota changed during 8 weeks of total meal replacement LED and determined their associations with host response in a sub-analysis of 211 overweight adults with pre-diabetes participating in the large multicentre PREVIEW (PREVention of diabetes through lifestyle intervention and population studies In Europe and around the World) clinical trial. METHODS Microbial community composition was analysed by Illumina sequencing of the hypervariable V3-V4 regions of the 16S ribosomal RNA (rRNA) gene. Butyrate production capacity was estimated by qPCR targeting the butyryl-CoA:acetate CoA-transferase gene. Bioinformatics and statistical analyses, such as comparison of alpha and beta diversity measures, correlative and differential abundances analysis, were undertaken on the 16S rRNA gene sequences of 211 paired (pre- and post-LED) samples as well as their integration with the clinical, biomedical and dietary datasets for predictive modelling. RESULTS The overall composition of the gut microbiota changed markedly and consistently from pre- to post-LED (P = 0.001), along with increased richness and diversity (both P < 0.001). Following the intervention, the relative abundance of several genera previously associated with metabolic improvements (e.g., Akkermansia and Christensenellaceae R-7 group) was significantly increased (P < 0.001), while flagellated Pseudobutyrivibrio, acetogenic Blautia and Bifidobacterium spp. were decreased (all P < 0.001). Butyrate production capacity was reduced (P < 0.001). The changes in microbiota composition and predicted functions were significantly associated with body weight loss (P < 0.05). Baseline gut microbiota features were able to explain ~25% of variation in total body fat change (post-pre-LED). CONCLUSIONS The gut microbiota and individual taxa were significantly influenced by the LED intervention and correlated with changes in total body fat and body weight in individuals with overweight and pre-diabetes. Despite inter-individual variation, the baseline gut microbiota was a strong predictor of total body fat change during the energy restriction period. TRIAL REGISTRATION The PREVIEW trial was prospectively registered at ClinicalTrials.gov ( NCT01777893 ) on January 29, 2013.
-
5.
Associations of quantity and quality of carbohydrate sources with subjective appetite sensations during 3-year weight-loss maintenance: Results from the PREVIEW intervention study.
Zhu, R, Larsen, TM, Poppitt, SD, Silvestre, MP, Fogelholm, M, Jalo, E, Hätönen, KA, Huttunen-Lenz, M, Taylor, MA, Simpson, L, et al
Clinical nutrition (Edinburgh, Scotland). 2022;(1):219-230
Abstract
BACKGROUND & AIMS The association of quantity and quality of carbohydrate sources with appetite during long-term weight-loss maintenance (WLM) after intentional weight loss (WL) is unclear. We aimed to investigate longitudinal associations of quantity and quality of carbohydrate sources with changes in subjective appetite sensations during WLM. METHODS This secondary analysis evaluated longitudinal data from the 3-year WLM phase of the PREVIEW study, a 2 × 2 factorial (diet-physical activity arms), multi-center, randomized trial. 1279 individuals with overweight or obesity and prediabetes (25-70 years; BMI≥25 kg m-2) were included. Individuals were merged into 1 group to assess longitudinal associations of yearly changes in appetite sensations. Quantity and quality of carbohydrate sources including total carbohydrate, glycemic index (GI), glycemic load (GL), and total dietary fiber were assessed via 4-day food diaries at 4 timepoints (26, 52, 104, and 156 weeks) during WLM. Visual analog scales were used to assess appetite sensations in the previous week. RESULTS During WLM, participants consumed on average 160.6 (25th, 75th percentiles 131.1, 195.8) g·day-1 of total carbohydrate, with GI 53.8 (48.7, 58.8) and GL 85.3 (67.2, 108.9) g day-1, and 22.3 (17.6, 27.3) g·day-1 of dietary fiber. In the available-case analysis, multivariable-adjusted linear mixed models with repeated measures showed that each 30-g increment in total carbohydrate was associated with increases in hunger (1.36 mm year-1, 95% CI 0.77, 1.95, P < 0.001), desire to eat (1.10 mm year-1, 0.59, 1.60, P < 0.001), desire to eat something sweet (0.99 mm year-1, 0.30, 1.68, P = 0.005), and weight regain (0.20%·year-1, 0.03, 0.36, P = 0.022). Increasing GI was associated with weight regain, but not associated with increases in appetite sensations. Each 20-unit increment in GL was associated with increases in hunger (0.92 mm year-1, 0.33, 1.51, P = 0.002), desire to eat (1.12 mm year-1, 0.62, 1.62, P < 0.001), desire to eat something sweet (1.13 mm year-1, 0.44, 1.81, P < 0.001), and weight regain (0.35%·year-1, 0.18, 0.52, P < 0.001). Surprisingly, dietary fiber was also associated with increases in desire to eat, after adjustment for carbohydrate or GL. CONCLUSIONS In participants with moderate carbohydrate and dietary fiber intake, and low to moderate GI, we found that higher total carbohydrate, GL, and total fiber, but not GI, were associated with increases in subjective desire to eat or hunger over 3 years. This study was registered as ClinicalTrials.gov, NCT01777893.
-
6.
Age- and sex-specific effects of a long-term lifestyle intervention on body weight and cardiometabolic health markers in adults with prediabetes: results from the diabetes prevention study PREVIEW.
Zhu, R, Craciun, I, Bernhards-Werge, J, Jalo, E, Poppitt, SD, Silvestre, MP, Huttunen-Lenz, M, McNarry, MA, Stratton, G, Handjiev, S, et al
Diabetologia. 2022;(8):1262-1277
-
-
Free full text
-
Abstract
AIMS/HYPOTHESIS Lifestyle interventions are the first-line treatment option for body weight and cardiometabolic health management. However, whether age groups or women and men respond differently to lifestyle interventions is under debate. We aimed to examine age- and sex-specific effects of a low-energy diet (LED) followed by a long-term lifestyle intervention on body weight, body composition and cardiometabolic health markers in adults with prediabetes (i.e. impaired fasting glucose and/or impaired glucose tolerance). METHODS This observational study used longitudinal data from 2223 overweight participants with prediabetes in the multicentre diabetes prevention study PREVIEW. The participants underwent a LED-induced rapid weight loss (WL) period followed by a 3 year lifestyle-based weight maintenance (WM) intervention. Changes in outcomes of interest in prespecified age (younger: 25-45 years; middle-aged: 46-54 years; older: 55-70 years) or sex (women and men) groups were compared. RESULTS In total, 783 younger, 319 middle-aged and 1121 older adults and 1503 women and 720 men were included in the analysis. In the available case and complete case analyses, multivariable-adjusted linear mixed models showed that younger and older adults had similar weight loss after the LED, whereas older adults had greater sustained weight loss after the WM intervention (adjusted difference for older vs younger adults -1.25% [95% CI -1.92, -0.58], p<0.001). After the WM intervention, older adults lost more fat-free mass and bone mass and had smaller improvements in 2 h plasma glucose (adjusted difference for older vs younger adults 0.65 mmol/l [95% CI 0.50, 0.80], p<0.001) and systolic blood pressure (adjusted difference for older vs younger adults 2.57 mmHg [95% CI 1.37, 3.77], p<0.001) than younger adults. Older adults had smaller decreases in fasting and 2 h glucose, HbA1c and systolic blood pressure after the WM intervention than middle-aged adults. In the complete case analysis, the above-mentioned differences between middle-aged and older adults disappeared, but the direction of the effect size did not change. After the WL period, compared with men, women had less weight loss (adjusted difference for women vs men 1.78% [95% CI 1.12, 2.43], p<0.001) with greater fat-free mass and bone mass loss and smaller improvements in HbA1c, LDL-cholesterol and diastolic blood pressure. After the WM intervention, women had greater fat-free mass and bone mass loss and smaller improvements in HbA1c and LDL-cholesterol, while they had greater improvements in fasting glucose, triacylglycerol (adjusted difference for women vs men -0.08 mmol/l [-0.11, -0.04], p<0.001) and HDL-cholesterol. CONCLUSIONS/INTERPRETATION Older adults benefited less from a lifestyle intervention in relation to body composition and cardiometabolic health markers than younger adults, despite greater sustained weight loss. Women benefited less from a LED followed by a lifestyle intervention in relation to body weight and body composition than men. Future interventions targeting older adults or women should take prevention of fat-free mass and bone mass loss into consideration. CLINICAL TRIAL REGISTRATION NUMBER ClinicalTrials.gov NCT01777893.
-
7.
Does the Effect of a 3-Year Lifestyle Intervention on Body Weight and Cardiometabolic Health Differ by Prediabetes Metabolic Phenotype? A Post Hoc Analysis of the PREVIEW Study.
Zhu, R, Jalo, E, Silvestre, MP, Poppitt, SD, Handjieva-Darlenska, T, Handjiev, S, Huttunen-Lenz, M, Mackintosh, K, Stratton, G, Navas-Carretero, S, et al
Diabetes care. 2022;(11):2698-2708
Abstract
OBJECTIVE To examine whether the effect of a 3-year lifestyle intervention on body weight and cardiometabolic risk factors differs by prediabetes metabolic phenotype. RESEARCH DESIGN AND METHODS This post hoc analysis of the multicenter, randomized trial, PREVention of diabetes through lifestyle interventions and population studies In Europe and around the World (PREVIEW), included 1,510 participants with prediabetes (BMI ≥25 kg ⋅ m-2; defined using oral glucose tolerance tests). Of these, 58% had isolated impaired fasting glucose (iIFG), 6% had isolated impaired glucose tolerance (iIGT), and 36% had IFG+IGT; 73% had normal hemoglobin A1c (HbA1c; <39 mmol ⋅ mol-1) and 25% had intermediate HbA1c (39-47 mmol ⋅ mol-1). Participants underwent an 8-week diet-induced rapid weight loss, followed by a 148-week lifestyle-based weight maintenance intervention. Linear mixed models adjusted for intervention arm and other confounders were used. RESULTS In the available-case and complete-case analyses, participants with IFG+IGT had greater sustained weight loss after lifestyle intervention (adjusted mean at 156 weeks -3.5% [95% CI, -4.7%, -2.3%]) than those with iIFG (mean -2.5% [-3.6%, -1.3%]) relative to baseline (P = 0.011). Participants with IFG+IGT and iIFG had similar cardiometabolic benefits from the lifestyle intervention. The differences in cardiometabolic benefits between those with iIGT and IFG+IGT were minor or inconsistent in different analyses. Participants with normal versus intermediate HbA1c had similar weight loss over 3 years and minor differences in cardiometabolic benefits during weight loss, whereas those with normal HbA1c had greater improvements in fasting glucose, 2-h glucose (adjusted between-group difference at 156 weeks -0.54 mmol ⋅ L-1 [95% CI -0.70, -0.39], P < 0.001), and triglycerides (difference -0.07 mmol ⋅ L-1 [-0.11, -0.03], P < 0.001) during the lifestyle intervention. CONCLUSIONS Individuals with iIFG and IFG+IGT had similar improvements in cardiometabolic health from a lifestyle intervention. Those with normal HbA1c had greater improvements than those with intermediate HbA1c.
-
8.
Forming new health behavior habits during weight loss maintenance-The PREVIEW study.
Huttunen-Lenz, M, Hansen, S, Raben, A, Westerterp-Plantenga, M, Macdonald, I, Stratton, G, Swindell, N, Martinez, JA, Handjieva-Darlenska, T, Poppitt, SD, et al
Health psychology : official journal of the Division of Health Psychology, American Psychological Association. 2022;(8):549-558
Abstract
INTRODUCTION Changing lifestyle habits to achieve and maintain weight loss can be effective in prevention of Type II diabetes. Ability to resist temptations is considered one of the key factors in behavior change. This study examined how habit strength, motivation, and temptations for an energy-dense diet developed during the maintenance stage of a behavior modification intervention tool. METHOD Participants with prediabetes and overweight/obesity were recruited in the two-phase trial PREVIEW with the aim to achieve ≥ 8% body weight loss over 2 months and maintain weight loss over a subsequent 34-month period. The four-stage intervention (PREVIEW Behavior Modification Intervention Toolbox, or PREMIT) supported participants in weight maintenance. Uni- and multivariate analyses were completed from the beginning of the PREMIT maintenance stage (Week 26 of the PREVIEW trial) with 962 individuals who completed the trial. RESULTS Habit strength and ability to resist temptations increased during the early PREMIT adherence stage (Weeks 26 to 52) before plateauing during middle (Weeks 52 to 104) and late (Weeks 104 to 156) PREMIT adherence stages. Higher habit strength for energy-dense diet was significantly associated with larger weight regain (p ≤ .007). No changes in motivation or interactions with PREMIT attendance were observed. DISCUSSION Changing diet habits is a complex, multifactorial process, with participants struggling at least with some aspects of weight maintenance. Habits against consuming energy-dense, sweet, and fatty food appeared effective in protecting against weight regain. The observed effect sizes were small, reflecting the complexity of breaking old habits and forming new ones to support long-term maintenance of weight loss. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
-
9.
The PREVIEW intervention study: Results from a 3-year randomized 2 x 2 factorial multinational trial investigating the role of protein, glycaemic index and physical activity for prevention of type 2 diabetes.
Raben, A, Vestentoft, PS, Brand-Miller, J, Jalo, E, Drummen, M, Simpson, L, Martinez, JA, Handjieva-Darlenska, T, Stratton, G, Huttunen-Lenz, M, et al
Diabetes, obesity & metabolism. 2021;23(2):324-337
-
-
-
Free full text
-
Plain language summary
The main risk for the development of type 2 diabetes is being overweight or obese. Strategies to decrease weight are important to prevent its development or reverse disease. This long-term, randomised control trial of 2326 adults with prediabetes, aimed to compare the effectiveness of different maintenance diets, after initial weight loss; one which was high protein and low glycaemic index, and the other which was moderate protein and moderate glycaemic index. These diets were then combined with either moderate intensity or high intensity exercise. The results showed that after 3 years, the incidence of type 2 diabetes was low and did not differ between the diet and exercise groups. However, more individuals achieved normal blood sugar levels when on a moderate protein diet combined with moderate exercise and when on a high protein diet combined with moderate exercise. The high protein diet with high intensity exercise was the least effective at maintaining normal blood sugar levels. The amount of weight lost was the same no matter what combination of diet and exercise. It was concluded that the incidence of type 2 diabetes was lower than expected with the diet and exercise regimes and did not differ between the diets. This study could be used by healthcare professionals to introduce a long-term protocol combining weight loss, healthy eating, and physical activity to pre diabetic individuals who want to reduce their chance of developing type 2 diabetes.
Abstract
AIM: To compare the impact of two long-term weight-maintenance diets, a high protein (HP) and low glycaemic index (GI) diet versus a moderate protein (MP) and moderate GI diet, combined with either high intensity (HI) or moderate intensity physical activity (PA), on the incidence of type 2 diabetes (T2D) after rapid weight loss. MATERIALS AND METHODS A 3-year multicentre randomized trial in eight countries using a 2 x 2 diet-by-PA factorial design was conducted. Eight-week weight reduction was followed by a 3-year randomized weight-maintenance phase. In total, 2326 adults (age 25-70 years, body mass index ≥ 25 kg/m2 ) with prediabetes were enrolled. The primary endpoint was 3-year incidence of T2D analysed by diet treatment. Secondary outcomes included glucose, insulin, HbA1c and body weight. RESULTS The total number of T2D cases was 62 and the cumulative incidence rate was 3.1%, with no significant differences between the two diets, PA or their combination. T2D incidence was similar across intervention centres, irrespective of attrition. Significantly fewer participants achieved normoglycaemia in the HP compared with the MP group (P < .0001). At 3 years, normoglycaemia was lowest in HP-HI (11.9%) compared with the other three groups (20.0%-21.0%, P < .05). There were no group differences in body weight change (-11% after 8-week weight reduction; -5% after 3-year weight maintenance) or in other secondary outcomes. CONCLUSIONS Three-year incidence of T2D was much lower than predicted and did not differ between diets, PA or their combination. Maintaining the target intakes of protein and GI over 3 years was difficult, but the overall protocol combining weight loss, healthy eating and PA was successful in markedly reducing the risk of T2D. This is an important clinically relevant outcome.
-
10.
Dose-Dependent Associations of Dietary Glycemic Index, Glycemic Load, and Fiber With 3-Year Weight Loss Maintenance and Glycemic Status in a High-Risk Population: A Secondary Analysis of the Diabetes Prevention Study PREVIEW.
Zhu, R, Larsen, TM, Fogelholm, M, Poppitt, SD, Vestentoft, PS, Silvestre, MP, Jalo, E, Navas-Carretero, S, Huttunen-Lenz, M, Taylor, MA, et al
Diabetes care. 2021;(7):1672-1681
-
-
Free full text
-
Abstract
OBJECTIVE To examine longitudinal and dose-dependent associations of dietary glycemic index (GI), glycemic load (GL), and fiber with body weight and glycemic status during 3-year weight loss maintenance (WLM) in adults at high risk of type 2 diabetes. RESEARCH DESIGN AND METHODS In this secondary analysis we used pooled data from the PREVention of diabetes through lifestyle Intervention and population studies in Europe and around the World (PREVIEW) randomized controlled trial, which was designed to test the effects of four diet and physical activity interventions. A total of 1,279 participants with overweight or obesity (age 25-70 years and BMI ≥25 kg ⋅ m-2) and prediabetes at baseline were included. We used multiadjusted linear mixed models with repeated measurements to assess longitudinal and dose-dependent associations by merging the participants into one group and dividing them into GI, GL, and fiber tertiles, respectively. RESULTS In the available-case analysis, each 10-unit increment in GI was associated with a greater regain of weight (0.46 kg ⋅ year-1; 95% CI 0.23, 0.68; P < 0.001) and increase in HbA1c. Each 20-unit increment in GL was associated with a greater regain of weight (0.49 kg ⋅ year-1; 0.24, 0.75; P < 0.001) and increase in HbA1c. The associations of GI and GL with HbA1c were independent of weight change. Compared with those in the lowest tertiles, participants in the highest GI and GL tertiles had significantly greater weight regain and increases in HbA1c. Fiber was inversely associated with increases in waist circumference, but the associations with weight regain and glycemic status did not remain robust in different analyses. CONCLUSIONS Dietary GI and GL were positively associated with weight regain and deteriorating glycemic status. Stronger evidence on the role of fiber is needed.