Correction: Development and validation of a model for predicting incident type 2 diabetes using quantitative clinical data and a Bayesian logistic model: A nationwide cohort and modeling study.

[This corrects the article DOI: 10.1371/journal.pmed.1003232.].

Development and validation of a model for predicting incident type 2 diabetes using quantitative clinical data and a Bayesian logistic model: A nationwide cohort and modeling study.

BACKGROUND: Obesity is closely related to the development of insulin resistance and type 2 diabetes (T2D). The prevention of T2D has become imperative to stem the rising rates of this disease. Weight loss is highly effective in preventing T2D; however, the at-risk pool is large, and a clinically meaningful metric for risk stratification to guide interventions remains a challenge. The objective of this study is to predict T2D risk using full-information continuous analysis of nationally sampled data from white and black American adults age ≥45 years. METHODS AND FINDINGS: A sample of 12,043 black (33%) and white individuals from a population-based cohort, REasons for Geographic And Racial Differences in Stroke (REGARDS) (enrolled 2003-2007), was observed through 2013-2016. The mean participant age was 63.12 ± 8.62 years, and 43.7% were male. Mean BMI was 28.55 ± 5.61 kg/m2. Risk factors for T2D regularly recorded in the primary care setting were used to evaluate future T2D risk using Bayesian logistic regression. External validation was performed using 9,710 participants (19% black) from Atherosclerotic Risk in Communities (ARIC) (enrolled 1987-1989), observed through 1996-1998. The mean participant age in this cohort was 53.86 ± 5.65 years, and 44.6% were male. Mean BMI was 27.15 ± 4.92 kg/m2. Predictive performance was assessed using the receiver operating characteristic (ROC) curves and area under the curve (AUC) statistics. The primary outcome was incident T2D. By 2016 in REGARDS, there were 1,602 incident cases of T2D. Risk factors used to predict T2D progression included age, sex, race, BMI, triglycerides, high-density lipoprotein, blood pressure, and blood glucose. The Bayesian logistic model (AUC = 0.79) outperformed the Framingham risk score (AUC = 0.76), the American Diabetes Association risk score (AUC = 0.64), and a cardiometabolic disease system (using Adult Treatment Panel III criteria) (AUC = 0.75). Validation in ARIC was robust (AUC = 0.85). Main limitations include the limited generalizability of the REGARDS sample to black and white, older Americans, and no time to diagnosis for T2D. CONCLUSIONS: Our results show that a Bayesian logistic model using full-information continuous predictors has high predictive discrimination, and can be used to quantify race- and sex-specific T2D risk, providing a new, powerful predictive tool. This tool can be used for T2D prevention efforts including weight loss therapy by allowing clinicians to target high-risk individuals in a manner that could be used to optimize outcomes.

Weight variability during self-monitored weight loss predicts future weight loss outcome.

BACKGROUND: Obesity treatments often do not produce long-term results. It is therefore critical to better understand biological and behavioral correlates or predictors of future weight change. OBJECTIVE: We tested the hypothesis that greater weight variability, independent of total body weight change, during early weight loss would predict degree of long-term success. SUBJECTS/METHODS: We included 24,009 American users of the Withings smart scale with over a year's worth of self-monitored weight data. Multilevel modeling was used to calculate weight variability as the root mean square error around participants' weight trajectory regression line, using weekly average weights from the first 12 weeks of weight loss. Linear regressions were then used to examine whether weight variability predicted weight change from week 12 to week 48, 72, and 96. RESULTS: Greater weight variability predicted less weight loss/more weight regain at week 48 (b ± SE: 1.18 ± 0.17, p < 0.001), week 72 (b ± SE: 1.45 ± 0.21, p < 0.001), and week 96 (b ± SE: 1.45 ± 0.23, p < 0.001), controlling for baseline BMI and overall weight change during the first 12 weeks. An interaction effect was found between weight variability and baseline BMI such that the relationship between weight variability and later weight change was stronger in individuals with lower baseline BMI. CONCLUSIONS: This study found that in a large population sample, weight variability early on during weight loss significantly predicted longer term weight loss outcomes. The results provide further support that weight variability be considered an important predictor of future weight change. Research is needed to understand the mechanisms underlying this effect.

Lack of negative autocorrelations of daily food intake on successive days challenges the concept of the regulation of body weight in humans.

According to most theories, the amount of food consumed on one day should be negatively related to intake on subsequent days. Several studies have observed such a negative correlation between the amount consumed on one day and the amount consumed two to four days later. The present study attempted to replicate this observation by re-examining data from a previous study where all food ingested over a 30-day observation period was measured. Nine male and seven female participants received a vegan diet prepared, dispensed, and measured in a metabolic unit. Autocorrelations were performed on total food intake consume on one day and that consumed one to five days later. A significant positive correlation was detected between the weight of food eaten on one day and on the amount consumed on the following day (r = 0.29, 95% CI [0.37, 0.20]). No correlation was found between weights of food consumed on one day and up to twelve days later (r = 0.09, 95% CI [0.24, -0.06]), (r = 0.11, 95% CI [0.26, -0.0.26]) (r = 0.02, 95% CI [0.15, -0.7]) (r = -0.08, 95% CI [0.11, -0.09]). The same positive correlation with the previous day's intake was observed at the succeeding breakfast but not at either lunch or dinner. However, the participants underestimated their daily energy need resulting in a small, but statistically significant weight loss. Daily food intake increased slightly (13 g/day), but significantly, across the 30-day period. An analysis of the previous studies revealed that the negative correlations observed by others was caused by a statistical artifact resulting from normalizing data before testing for the correlations. These results, when combined with the published literature, indicate that there is little evidence that humans precisely compensate for the previous day's intake by altering the amount consumed on subsequent days. Moreover, the small but persistent increase in food intake suggests that physiological mechanisms that affect food intake operate more subtly and over much longer periods of time than the meal or even total daily intake.

Cardiometabolic Disease Staging and Major Adverse Cardiovascular Event Prediction in 2 Prospective Cohorts.

BACKGROUND: Cardiometabolic risk prediction models that incorporate metabolic syndrome traits to predict cardiovascular outcomes may help identify high-risk populations early in the progression of cardiometabolic disease. OBJECTIVES: The purpose of this study was to examine whether a modified cardiometabolic disease staging (CMDS) system, a validated diabetes prediction model, predicts major adverse cardiovascular events (MACE). METHODS: We developed a predictive model using data accessible in clinical practice [fasting glucose, blood pressure, body mass index, cholesterol, triglycerides, smoking status, diabetes status, hypertension medication use] from the REGARDS (REasons for Geographic And Racial Differences in Stroke) study to predict MACE [cardiovascular death, nonfatal myocardial infarction, and/or nonfatal stroke]. Predictive performance was assessed using receiver operating characteristic curves, mean squared errors, misclassification, and area under the curve (AUC) statistics. RESULTS: Among 20,234 REGARDS participants with no history of stroke or myocardial infarction (mean age 64 ± 9.3 years, 58% female, 41% non-Hispanic Black, and 18% diabetes), 2,695 developed incident MACE (13.3%) during a median 10-year follow-up. The CMDS development model in REGARDS for MACE had an AUC of 0.721. Our CMDS model performed similarly to both the ACC/AHA 10-year risk estimate (AUC 0.721 vs 0.716) and the Framingham risk score (AUC 0.673). CONCLUSIONS: The CMDS predicted the onset of MACE with good predictive ability and performed similarly or better than 2 commonly known cardiovascular disease prediction risk tools. These data underscore the importance of insulin resistance as a cardiovascular disease risk factor and that CMDS can be used to identify individuals at high risk for progression to cardiovascular disease.

Effect of semaglutide 2.4 mg once weekly on 10-year type 2 diabetes risk in adults with overweight or obesity.

OBJECTIVE: In the Semaglutide Treatment Effect in People with obesity (STEP) trials, once-weekly subcutaneous semaglutide 2.4 mg plus lifestyle intervention reduced body weight and improved cardiometabolic parameters in adults with obesity (or overweight with weight-related comorbidities). Effects on the risk of developing type 2 diabetes (T2D) require investigation. METHODS: STEP 1 (68 weeks) and 5 (104 weeks) randomized participants to semaglutide 2.4 mg or placebo. STEP 4 included a 20-week semaglutide run-in followed by randomization to 48 weeks of continued semaglutide or withdrawal (placebo). Ten-year T2D risk scores were calculated post hoc using Cardiometabolic Disease Staging. RESULTS: In STEP 1 (N = 1583), relative risk score reductions were greater with semaglutide versus placebo (semaglutide: -61.1%; placebo: -12.9%; p < 0.0001). These reductions were maintained to week 104 in STEP 5 (N = 295; semaglutide: -60.0%; placebo: 3.5%; p < 0.0001). Risk scores during the STEP 4 run-in period (N = 776) were reduced from 20.6% to 11.1% and further to 7.7% at week 68 with continued semaglutide, increasing to 15.4% with withdrawal (relative risk score change: semaglutide: -32.1%; placebo: +40.6%; p < 0.0001). Risk score reductions mirrored weight loss. CONCLUSIONS: Cardiometabolic Disease Staging risk assessment suggests that once-weekly semaglutide 2.4 mg may substantially lower 10-year T2D risk in people with overweight or obesity.

Three-Year Follow-Up of Participants from a Self-Weighing Randomized Controlled Trial.

Frequent self-weighing is associated with weight loss maintenance. Several years ago, we investigated frequent self-weighing's effect on weight loss and found the participants lost a significant amount of weight. Three years after this trial's end, participants were contacted for an update on their weight and self-weighing frequency. Weight change and self-weighing frequency since the end of the study were assessed. We hypothesized that participants who maintained frequent self-weighing behavior would have maintained their weight loss. Out of 98 participants enrolled in the RCT, 37% (n = 36) participated in this follow-up study. Total weight loss during the trial for the follow-up participants was 12.7 ± 19.4 lbs (p < 0.001). Three years after intervention, participants regained 0.9 ± 4.34 lbs, a value that was not statistically different from zero (p = 0.75). This did not differ by gender (p = 0.655). Over 75% of these participants continued to weigh themselves at least once a week. Frequent self-weighing may be an effective, low-cost strategy for weight loss maintenance. Future research should further investigate the role of self-weighing in long-term weight gain prevention.