Time-to-event outcomes are common in medical research as they offer more information than simply whether or not an event occurred. To handle these outcomes, as well as censored observations where the event was not observed during follow-up, survival analysis methods should be used. Kaplan-Meier estimation can be used to create graphs of the observed survival curves, while the log-rank test can be used to compare curves from different groups. If it is desired to test continuous predictors or to test multiple covariates at once, survival regression models such as the Cox model or the accelerated failure time model (AFT) should be used. The choice of model should depend on whether or not the assumption of the model (proportional hazards for the Cox model, a parametric distribution of the event times for the AFT model) is met. The goal of this paper is to review basic concepts of survival analysis. Discussions relating the Cox model and the AFT model will be provided. The use and interpretation of the survival methods model are illustrated using an artificially simulated dataset.

No abstract

PurposeTo compare the effects of six weeks of high intensity interval training (HIIT) vs continuous moderate intensity training (MIT) for improving body composition, insulin sensitivity (SI), blood pressure, blood lipids, and cardiovascular fitness in a cohort of sedentary overweight or obese young men. We hypothesized that HIIT would result in similar improvements in body composition, cardiovascular fitness, blood lipids, and SI as compared to the MIT group, despite requiring only one hour of activity per week compared to five hours per week for the MIT group.Methods28 sedentary overweight or obese men (age, 20 ± 1.5 years, body mass index 29.5 ± 3.3 kg/m2) participated in a six week exercise treatment. Participants were randomly assigned to HIIT or MIT and evaluated at baseline and post-training. DXA was used to assess body composition, graded treadmill exercise test to measure cardiovascular fitness, oral glucose tolerance to measure SI, nuclear magnetic resonance spectroscopy to assess lipoprotein particles, and automatic auscultation to measure blood pressure.ResultsA greater improvement in VO2peak was observed in MIT compared to HIIT (11.1% vs 2.83%, P = 0.0185) in the complete-case analysis. No differences were seen in the intention to treat analysis, and no other group differences were observed. Both exercise conditions were associated with temporal improvements in % body fat, total cholesterol, medium VLDL, medium HDL, triglycerides, SI, and VO2peak (P < 0.05).ConclusionParticipation in HIIT or MIT exercise training displayed: 1) improved SI, 2) reduced blood lipids, 3) decreased % body fat, and 4) improved cardiovascular fitness. While both exercise groups led to similar improvements for most cardiometabolic risk factors assessed, MIT led to a greater improvement in overall cardiovascular fitness. Overall, these observations suggest that a relatively short duration of either HIIT or MIT training may improve cardiometabolic risk factors in previously sedentary overweight or obese young men, with no clear advantage between these two specific regimes (Clinical Trial Registry number NCT01935323).Trial RegistrationClinicalTrials.gov NCT01935323

Introduction Much is to be learned about humanbreast milk (HBM) Purpose Extend our knowledge of HBM by investigating the role of maternal body mass index (BMI), sex and stage of lactation (month 1 vs. 6) on HBM insulin, glucose, leptin, IL-6 and TNF-α and their associations with infant body composition. Methods Thirty-seven exclusively breastfeeding infants (n=37; 16♀, 21♂) and their mothers (19–47 kg/m2) were studied at 1 and 6 months of lactation. Infants had body composition measured (using dual energy X-ray absorptiometry (DXA)) and HBM collected. Results A significant interaction between maternal BMI and infant sex on insulin levels (p = 0.0322) was observed, such that insulin was 229% higher in obese mothers nursing female infants than in normal weight mothers nursing female infants and 179% higher than obese mothers nursing male infants. For leptin, a significant association with BMI category was observed (p < 0.0001) such that overweight and obese mothers had 96.5% and 315.1% higher leptin levels than normal weight mothers, respectively. Leptin was also found to have a significant (p = 0.0004) 33.7% decrease from month 1 to month 6, controlling for BMI category and sex. A significant inverse relationship between month 1 leptin levels and infant length (p=0.0257), percent fat (p=0.0223), total fat mass (p=0.0226) and trunk fat mass (p=0.0111) at month 6 was also found. No associations or interactions were observed for glucose, TNF-α, or IL-6. Conclusions These data demonstrates that maternal BMI, infant sex and stage of lactation affect the compositional make-up of insulin and leptin.

We identify 10 common errors and problems in the statistical analysis, design, interpretation, and reporting of obesity research and discuss how they can be avoided. The 10 topics are: 1) misinterpretation of statistical significance, 2) inappropriate testing against baseline values, 3) excessive and undisclosed multiple testing and “p-value hacking,” 4) mishandling of clustering in cluster randomized trials, 5) misconceptions about nonparametric tests, 6) mishandling of missing data, 7) miscalculation of effect sizes, 8) ignoring regression to the mean, 9) ignoring confirmation bias, and 10) insufficient statistical reporting. We hope that discussion of these errors can improve the quality of obesity research by helping researchers to implement proper statistical practice and to know when to seek the help of a statistician.

Late-term thrombosis on drug-eluting stents is an emerging problem that might be addressed using extremely thin, biologically-active hydrogel coatings. We report a dip-coating strategy to covalently link poly(ethylene glycol) (PEG) to substrates, producing coatings with <≈100 nm thickness. Gelation of PEG-octavinylsulfone with amines in either bovine serum albumin (BSA) or PEG-octaamine was monitored by dynamic light scattering (DLS), revealing the presence of microgels before macrogelation. NMR also revealed extremely high end group conversions prior to macrogelation, consistent with the formation of highly crosslinked microgels and deviation from Flory-Stockmayer theory. Before macrogelation, the reacting solutions were diluted and incubated with nucleophile-functionalized surfaces. Using optical waveguide lightmode spectroscopy (OWLS) and quartz crystal microbalance with dissipation (QCM-D), we identified a highly hydrated, protein-resistant layer with a thickness of approximately 75 nm. Atomic force microscopy in buffered water revealed the presence of coalesced spheres of various sizes but with diameters less than about 100 nm. Microgel-coated glass or poly(ethylene terephthalate) exhibited reduced protein adsorption and cell adhesion. Cellular interactions with the surface could be controlled by using different proteins to cap unreacted vinylsulfone groups within the coating.

Deriving statistical models to predict one variable from one or more other variables, or predictive modeling, is an important activity in obesity and nutrition research. To determine the quality of the model, it is necessary to quantify and report the predictive validity of the derived models. Conducting validation of the predictive measures provides essential information to the research community about the model. Unfortunately, many articles fail to account for the nearly inevitable reduction in predictive ability that occurs when a model derived on one dataset is applied to a new dataset. Under some circumstances, the predictive validity can be reduced to nearly zero. In this overview, we explain why reductions in predictive validity occur, define the metrics commonly used to estimate the predictive validity of a model (e.g., R2, mean squared error, sensitivity, specificity, receiver operating characteristic, concordance index), and describe methods to estimate the predictive validity (e.g., cross-validation, bootstrap, adjusted and shrunken R2). We emphasize that methods for estimating the expected reduction in predictive ability of a model in new samples are available and this expected reduction should always be reported when new predictive models are introduced.

Given the increasing evidence that supports the ability of humans to taste non-esterified fatty acids (NEFA), recent studies have sought to determine if relationships exist between oral sensitivity to NEFA (measured as thresholds), food intake and obesity. Published findings suggest there is either no association or an inverse association. A systematic review and meta-analysis was conducted to determine if differences in fatty acid taste sensitivity or intensity ratings exist between individuals who are lean or obese. A total of 7 studies that reported measurement of taste sensations to non-esterified fatty acids by psychophysical methods (e.g.,studies using model systems rather than foods, detection thresholds as measured by a 3-alternative forced choice ascending methodology were included in the meta-analysis. Two other studies that measured intensity ratings to graded suprathreshold NEFA concentrations were evaluated qualitatively. No significant differences in fatty acid taste thresholds or intensity were observed. Thus, differences in fatty acid taste sensitivity do not appear to precede or result from obesity.