Correction: A novel scaling methodology to reduce the biases associated with missing data from commercial activity monitors.

[This corrects the article DOI: 10.1371/journal.pone.0235144.].

A novel scaling methodology to reduce the biases associated with missing data from commercial activity monitors.

BACKGROUND: Commercial physical activity monitors have wide utility in the assessment of physical activity in research and clinical settings, however, the removal of devices results in missing data and has the potential to bias study conclusions. This study aimed to evaluate methods to address missingness in data collected from commercial activity monitors. METHODS: This study utilised 1526 days of near complete data from 109 adults participating in a European weight loss maintenance study (NoHoW). We conducted simulation experiments to test a novel scaling methodology (NoHoW method) and alternative imputation strategies (overall/individual mean imputation, overall/individual multiple imputation, Kalman imputation and random forest imputation). Methods were compared for hourly, daily and 14-day physical activity estimates for steps, total daily energy expenditure (TDEE) and time in physical activity categories. In a second simulation study, individual multiple imputation, Kalman imputation and the NoHoW method were tested at different positions and quantities of missingness. Equivalence testing and root mean squared error (RMSE) were used to evaluate the ability of each of the strategies relative to the true data. RESULTS: The NoHoW method, Kalman imputation and multiple imputation methods remained statistically equivalent (p<0.05) for all physical activity metrics at the 14-day level. In the second simulation study, RMSE tended to increase with increased missingness. Multiple imputation showed the smallest RMSE for Steps and TDEE at lower levels of missingness (<19%) and the Kalman and NoHoW methods were generally superior for imputing time in physical activity categories. CONCLUSION: Individual centred imputation approaches (NoHoW method, Kalman imputation and individual Multiple imputation) offer an effective means to reduce the biases associated with missing data from activity monitors and maximise data retention.

Processed meat intake and incidence of colorectal cancer: a systematic review and meta-analysis of prospective observational studies.

The objective was to use accumulated evidence to explore the association between processed meat intake and risk of colorectal cancer (CRC) and to investigate the reliability of associations by evaluating patterns of risk by study population characteristics and research quality parameters. We included 29 observational prospective cohort studies with relative risk estimates and 95% confidence intervals for CRC according to various levels of processed meat consumption. Risk of bias was assessed using Risk Of Bias In Non-randomized Studies-of Interventions (ROBINS-I) tool. Data sources were PubMed and Embase up to January 2017. The summary relative risks for high versus low processed meat consumption and risk of CRC, colon, and rectal cancer were 1.13 (95% CI: 1.01, 1.26), 1.19 (95% CI: 1.09, 1.31), and 1.21 (95% CI: 0.98, 1.49), respectively. Similar estimates were observed for the dose-response analyses. Heterogeneity across studies was detected in most analytical models. The overall judgment showed that two out of 29 studies had a moderate risk of bias, 25 had a serious risk of bias, and 2 had a critical risk of bias. The bias domains most often rated critical were bias due to risk of confounding, bias due to missing data, and selective outcome reporting bias. Although this meta-analysis indicates a modest association between processed meat intake and an increased risk of CRC, our assessment of internal validity warrants a cautious interpretation of these results, as most of the included studies were judged to have serious or critical risks of bias.

Prevalence and predictors of weight loss maintenance: a retrospective population-based survey of European adults with overweight and obesity.

BACKGROUND: The prevalence of weight loss attempts has increased worldwide, although the extent to which sustained weight loss is achieved is unknown. There is insufficient research into weight loss maintenance (WLM) in individuals with overweight or obesity who have recently lost clinically significant amounts of weight (≥5%), particularly in the European general population. The present study aimed to determine the prevalence and retrospective predictors of WLM in population-based samples of European adults with overweight or obesity who had made a recently completed weight loss attempt. METHODS: Participants (N = 2000) in UK, Denmark and Portugal completed an online survey about loss and regain in their most recent completed weight loss attempt, features of their attempt (duration, self-weighing, lapses, strategies), as well as loss of control and binge eating. Multiple regression analysis was used to determine factors retrospectively associated with WLM in those who achieved clinically significant weight loss (n = 1272). RESULTS: Mean (SD) self-reported weight loss was 9% (8%) and mean (SD) regain was 96.3% (9%) of participants' start weight. Twenty-three percent of the total sample had maintained weight loss of ≥5% for at least 1 month. Controlling for weight loss and time since attempt, predictors of better WLM were avoidance of a temporary lapse, infrequent/absent loss of control and binge eating, and use of a greater number of dietary strategies for WLM (r2  = 0.338, P < 0.001). PRINCIPAL CONCLUSIONS: Factors associated with recent successful WLM indicate the importance of the continued use of dietary and other strategies for WLM, particularly in the face of a lapse, as well as the need to manage dysfunctional eating behaviours.

Lower risk of hip fractures among Swedish women with large hips?

In women, a large hip circumference (HC) related to lower hip fracture risk, independent of age and regardless if HC was measured long before or closer to the fracture. In older women, body mass index (BMI) explained the protection. INTRODUCTION: In postmenopausal women, HC has been suggested to inversely associate with hip fracture while this has not been investigated in middle-aged women. We examined the association between HC, measured at two different time points, and hip fracture in a Swedish female population-based sample monitored for incident hip fractures over many years. METHODS: Baseline HC, measured in 1968 or 1974 (n = 1451, mean age 47.6 years), or the HC measures that were the most proximal before event or censoring (n = 1325, mean age 71.7 years), were used to assess the effects of HC on hip fracture risk in women participating in the Prospective Population Study of Women in Gothenburg. HC was parameterized as quintiles with the lowest quintile (Q1) as reference. Incident hip fractures over 45 years of follow-up (n = 257) were identified through hospital registers. RESULTS: Higher quintiles of HC at both baseline and proximal to event were inversely associated with hip fracture risk in age-adjusted models, but only baseline HC predicted hip fractures independently of BMI and other covariates (HR (95% CI) Q2, 0.85 (0.56-1.27); Q3, 0.59 (0.36-0.96); Q4, 0.57 (0.34-0.96); Q5, 0.58 (0.31-1.10)). CONCLUSIONS: A large HC is protective against hip fracture in midlife and in advanced age, but the association between proximal HC and hip fracture was explained by concurrent BMI suggesting that padding was not the main mechanism for the association. The independent protection seen in middle-aged women points to other mechanisms influencing bone strength.

Habitual coffee consumption and changes in measures of adiposity: a comprehensive study of longitudinal associations.

BACKGROUND: A relationship between change in coffee consumption and reduced long-term weight gain has been suggested, but current evidence is inconsistent. OBJECTIVE: To examine longitudinal associations between coffee consumption and changes in body mass index (BMI), fat mass index (FMI), fat-free mass index (FFMI), body fat percentage (BF %) and waist circumference (WC). DESIGN: The study consisted of 2128 participants from the Danish part of the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) cohort with repeated information on coffee consumption, adiposity measures and covariates during an 11-year period. Linear regression analyses were conducted to assess the associations between baseline coffee consumption and subsequent change in adiposity measures. The same analyses were conducted analyzing associations between change in coffee consumption and concurrent as well as subsequent changes in adiposity measures. RESULTS: We found no consistent evidence of associations between baseline coffee consumption and subsequent 6-year changes in adiposity measures. A statistically significant association between increased coffee consumption over a 6-year period and decreased concurrent gain in BMI, FMI, BF % and WC (-0.05 kg m-2 (95% confidence interval (CI): -0.07, -0.02), -0.04 kg m-2 (95% CI: -0.06, -0.02), -0.08% (95% CI: -0.13, -0.04) and -0.23 cm (95% CI: -0.34, -0.12), respectively, per 1 cup day-1 increase in coffee consumption) was found. No association was seen between change in coffee consumption and concurrent change in FFMI. Moreover, an initial change in coffee consumption during the first 5-year period was not associated with change in adiposity during the subsequent 6-year period. CONCLUSIONS: Increased coffee consumption was associated with a decreased concurrent gain in body weight, fat mass and waist circumference, but the associations were weak. Moreover, a causal relationship could not be established, as we found no evidence of associations between an initial change in coffee consumption and subsequent change in adiposity.

Shorter sleep duration is associated with higher energy intake and an increase in BMI z-score in young children predisposed to overweight.

BACKGROUND: Inadequate sleep has been shown to be a contributor to obesity in both children and adults. Less evidence is available for toddlers and among those with higher obesity risk. The objective of this study was to examine the relationship between sleep patterns and body weight development in a group of young obesity-predisposed children, and to assess whether intakes of energy or macronutrients mediate this relationship. METHODS: Participants included 368 Danish children aged 2-6 years from the Healthy Start Study, a 1.3 year randomised controlled intervention trial. Sleep habits were measured using a 7-day sleep diary. Multivariate linear regression with adjustment for confounders was used to assess the association of sleep duration and sleep variability with 1.3 year changes (Δ) in body mass index (BMI) z-score from baseline to follow-up. RESULTS: The average nighttime sleep duration was 10.7 h (range 8.8-12.5 h). After controlling for potential confounders, a significant inverse association between nighttime sleep duration and ΔBMI z-score (ß=-0.090, P=0.046) was observed. This relationship was mediated by energy intake, with all macronutrients contributing to this mediation effect. No associations were found for sleep variability and ΔBMI z-score but baseline intake of added sugars and sugary beverages were positively associated with sleep variability. CONCLUSION: Shorter sleep duration, mediated by energy intake in early in life, seems a risk factor for weight gain among young obesity-predisposed children.

The association between parental separation during childhood and obesity in adulthood: a Danish twin study.

OBJECTIVE: The purpose of this study was to examine if parental separation during childhood is associated with obesity in adulthood. METHODS: A co-twin case-control study of 146 adult same-sexed twin pairs with discordant body mass index (BMI) (i.e. one of the twins should have a BMI of 20-25 kg/m2, and the co-twin's BMI ≥ 30 kg/m2) was selected from Danish Twin Registry (DTR). In total of 236 eligible twin individuals participated in the study. Childhood parental separation (defined as separation from one of the biological parents, regardless of the reason for separation) for at least one year prior to age 17 was self-reported. The statistical analysis includes logistic and linear regression models using STATA 13.0. RESULTS: There were no differences in the odds of developing obesity in adulthood between the twin who stayed with a father and the co-twin who was separated from him for at least 1 year prior to age 17 [OR = 1.22, 95%CI (0.46-3.34), p = 0.65]. Separation from a mother also showed no differences in the odds for developing obesity [OR = 0.90, 95%CI (0.32-2.46), p = 0.82]. CONCLUSIONS: Because of the limited number of discordant twin pairs for childhood parental separation, we cannot provide evidence to suggest that separation from parents in childhood was associated with developing obesity in adulthood. Further studies of pooling discordant twins from several countries should be considered.

Vitamin D fortification and seasonality of birth in type 1 diabetic cases: D-tect study.

Fortification of margarine with vitamin D was mandatory in Denmark during 1961-1985. The aim of the study was to assess whether gestational and early infancy exposure to margarine fortification was associated with seasonality of birth in Danish type 1 diabetes (T1D) patients. The risks of T1D in Danes born during various exposure periods around margarine fortification termination in 1985 were analyzed. As expected, the T1D hazards in males unexposed to margarine fortification and born in spring were higher than in males born in autumn: relevant hazard ratios (95% confidence intervals) in various exposure groups ranged from 1.74 (1.112/2.708) to 37.43 (1.804/776.558). There were no indications of seasonality of birth in males exposed to fortification, nor in both exposed and unexposed females. The study suggests that early life exposure to low-dose vitamin D from fortified food eliminates seasonality of birth in T1D male patients. Further studies are required to investigate the identified gender differences.

Prenatal exposure to vitamin-D from fortified margarine and milk and body size at age 7 years.

BACKGROUND/OBJECTIVES: Prenatal vitamin-D deficiency may be associated with increased risk of obesity later in life. Using two national vitamin-D fortification programs as the setting for a societal experiment, we investigated whether exposure to vitamin-D from fortified margarine and low-fat milk during foetal life was associated with body size at 7 years of age. SUBJECTS/METHODS: Vitamin-D fortification of margarine was mandatory in Denmark from 1961 to 1985, and voluntary fortification of low-fat milk was permitted from 1972 to 1976. Using information on body mass index (BMI) Z-score at the age of 7 years of 54,270 children, who were measured during the mandatory Copenhagen School Health examination, we compared children according to whether the mothers were pregnant during the fortification programs or not. The comparisons were performed for children born just before and after initiation or termination of margarine and milk fortification periods, respectively. In total four sets of analyses were performed. RESULTS: We observed no difference in mean BMI Z-score between children exposed to vitamin-D fortification in utero and non-exposed children. Similar results were observed for overweight and obesity. CONCLUSIONS: Prenatal exposure to vitamin-D from fortification of margarine and low-fat milk showed no association with body size at 7 years.

Liquid versus solid energy intake in relation to body composition among Australian children.

BACKGROUND: The debate about whether energy consumed in liquid form is more obesogenic than energy consumed in solid form remains equivocal. We aimed to evaluate the effects of liquid versus solid energy intake and different beverage types on changes in childhood adiposity. METHODS: Our analyses included 8-year-old Australian children (n = 158) participating in the Childhood Asthma Prevention Study. Dietary information was collected using three 24-h recalls at age 9 years. Multivariate linear regression was used to evaluate the effects of liquid versus solid energy intake and different beverage types on changes in body mass index (BMI) Z-score from ages 8 to 11.5 years (△BMIz(8-11.5y)) and percentage body fat (%BF) at age 11.5 years (%BF(11.5y)). Substitution models were used to evaluate the effects of substituting other beverage types for sugar-sweetened beverages (SSB). RESULTS: Liquid energy intake (1 MJ day(-1)) was more closely associated with both △BMIz(8-11.5y) (ß = 0.23, P = 0.02) and %BF(11.5y) (ß = 2.31%, P = 0.01) than solid energy intake (△BMIz(8-11.5y): ß = 0.12, P = 0.01 and %BF(11.5y): ß = 0.80%, P = 0.07). SSB consumption (100 g day(-1)) was directly associated with △BMIz(8-11.5y) (ß = 0.08, P = 0.02) and %BF(11.5y) (ß = 0.92%, P = 0.004),whereas diet drinks (100 g day(-1) ) were inversely associated with △BMIz(8-11.5y) (ß = 0.18, P = 0.02). Substitution of 100 g of SSB by 100 g of water or diet drink, but not other beverages, was inversely associated with both △BMIz(8-11.5y) and %BF(11.5y) (P < 0.01). CONCLUSIONS: Our findings indicate that liquid energy is more obesogenic than solid energy. In particular, SSB, but not other beverage types, are a significant predictor of childhood adiposity and replacing SSB with water can have long-term beneficial effects on childhood adiposity.

Dietary protein and urinary nitrogen in relation to 6-year changes in fat mass and fat-free mass.

BACKGROUND: In contrast to the physiological expectation, observational studies show that greater protein intake is associated with subsequent body weight (BW) gain. An increase in fat-free mass (FFM) due to the anabolic effects of protein could explain this. OBJECTIVE: To examine associations between protein intake and subsequent changes in fat mass (FM) and FFM in longitudinal, observational data. DESIGN: A health examination, including measures of FM and FFM by bioelectrical impedance at baseline and follow-up 6 years later, was conducted. Diet history interviews (DHI) were performed, and 24-h urinary nitrogen collection at baseline was done. In total, 330 participants with DHI, of whom 227 had validated and complete 24-h urine collection data, were analyzed. Macronutrient energy substitution models were used. RESULTS: Mean estimated protein intake was 14.6 E% from DHI and 11.3 E% from urinary nitrogen. Estimated from DHI, FM increased 46 g per year, with every 1 E% protein substituted for fat (95% confidence interval (CI) = 13, 79; P = 0.006), and FFM increased 15 g per year (1, 30; P = 0.046). Results were similar in other substitution models. Estimated from urinary nitrogen, FM increased 53 g per year, with 1 E% protein substituted for other macronutrients (24, 81; P < 0.0005), and FFM increased 18 g per year (6, 31; P = 0.004). CONCLUSION: Within a habitual range, a greater protein intake was associated with BW gain, mostly in FM. This is in contrast to the expectations based on physiological and clinical trials, and calls for a better understanding of how habitual dietary protein influences long-term energy balance, versus how greater changes in dietary proteins may influence short-term energy balance.

A break in the obesity epidemic? Explained by biases or misinterpretation of the data?

Recent epidemiologic papers are presenting prevalence data suggesting breaks and decreases in obesity rates. However, before concluding that the obesity epidemic is not increasing anymore, the validity of the presented data should be discussed more thoroughly. We had a closer look into the literature presented in recent reviews to address the major potential biases and distortions, and to develop insights about how to interpret the presented suggestions for a potential break in the obesity epidemic. Decreasing participation rates, the use of reported rather than measured data and small sample sizes, or lack of representativeness, did not seem to explain presented breaks in the obesity epidemic. Further, available evidence does not suggest that stabilization of obesity rates is seen in higher socioeconomic groups only, or that urbanization could explain a potential break in the obesity epidemic. However, follow-ups of short duration may, in part, explain the apparent break or decrease in the obesity epidemic. On the other hand, a single focus on body mass index (BMI) ⩾25 or ⩾30 kg m(-)(2) is likely to mask a real increase in the obesity epidemic. And, in both children and adults, trends in waist circumferences were generally suggesting an increase, and were stronger than those reported for trends in BMI. Studies concluding that there is a recent break in the obesity epidemic need to be interpreted with caution. Reported studies presenting a break were mostly of short duration. Further, focusing on trends in waist circumference rather than BMI leads to a less optimistic conclusion: the public health problem of obesity is still increasing.

Sugar-sweetened beverages consumption in relation to changes in body fatness over 6 and 12 years among 9-year-old children: the European Youth Heart Study.

BACKGROUND/OBJECTIVES: In parallel with the obesity epidemic, consumption of sugar-sweetened beverages (SSB) has risen over the same period. Our aim was to investigate associations between the consumption of SSB in childhood and adolescence with subsequent changes in body fatness in early adulthood. SUBJECTS/METHODS: A longitudinal study of 9-year-old children (n=283) enrolled in the Danish part of the European Youth Heart Study with a 6-year and 12-year follow-up. Data were collected at ages 9, 15 and 21 years. Multivariate regression analyses with adjustment for potential confounders were used to evaluate the effect of SSB consumption at 9 and 15 years and change in SSB consumption from 9-15 years on subsequent change in body fatness until 21 years. RESULTS: Subjects who consumed more than one serve of SSB daily at age 15 years had larger increases in body mass index (BMI) (ß=0.92, P=0.046) and waist circumference (WC) (ß=2.69, P=0.04) compared to non-consumers over the subsequent 6 years. In addition, subjects who increased their SSB consumption from age 9-15 years also had larger increases in BMI (ß=0.91, P=0.09) and WC (ß=2.72, P=0.04) from 15-21 years, compared to those who reported no change in consumption. No significant association was observed from 9-21 years. CONCLUSION: This study provides new evidence that SSB consumption in adolescence and changes in SSB consumption from childhood to adolescence are both significant predictors of change in body fatness later in early adulthood.

Association between sweet drink intake and adiposity in Danish children participating in a long-term intervention study.

BACKGROUND: In some previous studies direct associations between intake of soft drinks, sugar-sweetened beverages and adiposity have been reported. The majority of these studies were, however, conducted in the USA and it is uncertain if the results are applicable to non-US countries. OBJECTIVE: To assess the association between sweet drink intake at age 6 and 9 years and the subsequent 3- to 7-year changes in body mass index (BMI) and sum of four skin-folds (Σ4SF). METHODS: Information on sweet drink intake (7 days food record) and physical activity (accelerometer) was obtained at age 6 years (n = 366) [Correction made here after initial online publication.] and 9 years (n = 269). Weight, height and Σ4SF were measured at age 6, 9 and 13 years. Additional information on socio-economic status, maternal BMI and pubertal status was obtained. RESULTS: No associations were observed between sweet drink intake at age 6 years and change in BMI or logΣ4SF from age 6 to 9 years or 6 to 13 years. Also, no associations were observed between change in sweet drink intake from age 6 to 9 years and subsequent change in BMI or logΣ4SF from age 9 to 13 years. A weak direct association was observed between sweet drink intake at age 9 years and change in logΣ4SF from age 9 to 13 years (per 100 g ∼ 3.38 fl oz) (ß: 0.014, 95% confidence interval [CI]: -0.001 to 0.029, P = 0.06), while no association was seen for BMI. In supplementary analyses a similar association was observed for soft drinks (ß: 0.087, 95% CI: 0.048 to 0.126, P = 0.001) but only in the intervention group. CONCLUSION: We observed associations between intake of sweet drinks and soft drinks and change in skin-fold thickness in a group of Danish children. However, as the associations did not remain significant when multiple testing was considered or was only significant among children from the intervention group, the results do not confirm or refute the direct association reported in previous studies.

Combined influence of leisure-time physical activity and hip circumference on all-cause mortality.

UNLABELLED: Hip circumference has been shown to be inversely associated with mortality. Muscle atrophy in the gluteofemoral region may be a possible explanation and thus physical activity is likely to play an important role. OBJECTIVE: To estimate the combined effects of hip circumference and physical activity on mortality. DESIGN AND METHODS: From the Copenhagen City Heart Study, 3,358 men and 4,350 women aged 21 to 93 years without pre-existing diagnosis of diabetes, stroke, ischemic heart disease, or cancer in 1991-1994 and with complete information on the variables of interest were included in the analyses. The participants were followed to 2009 in the Danish Civil Registration System, with 1.3% loss to follow-up and 2,513 deaths. Hazard ratios (HR) were estimated for combinations of physical activity and hip circumference. RESULTS: Hip circumference was inversely associated with mortality irrespective of being physically active or not. However, being physically active seemed to counterbalance some of the adverse health effects of a small hip circumference; when comparing inactive to active, the excess mortality at the 25th percentile of hip circumference is 40% in men (HR = 1.40, 95% CI: 1.14-1.72) and 33% in women (HR = 1.33, CI: 1.10-1.62). These associations were observed after adjustment for waist circumference and weight change in the 6 months before the examination. CONCLUSION: Less effects of physical activity were found in individuals with greater hip circumferences. A small hip circumference appears hazardous to survival. However, being physically active may counterbalance some of the hazardous effects of a small hip circumference.

The use of bioelectrical impedance analysis for body composition in epidemiological studies.

BACKGROUND/OBJECTIVES: Bioelectrical impedance analysis (BIA) is a relatively simple, inexpensive and non-invasive technique to measure body composition and is therefore suitable in field studies and larger surveys. SUBJECTS/METHODS: We performed an overview of BIA-derived body fat percentages (BF%) from 55 published studies of healthy populations aged 6-80 years. In addition, the relationship between body mass index (BMI) and body composition is documented in the context of BIA as a good alternative to closely differentiate which composition of the body better relates to the risk of cardiovascular diseases (CVDs)and all-cause mortality. RESULTS AND CONCLUSIONS: BIA-estimated percentage of BF varies greatly with population and age. BIA-estimated BF% is directly and closely related to various health outcomes such as CVDs, which is in contrast to BMI where both high and low BMIs are associated with increased risk of developing chronic diseases. Studies, among others using BIA, suggest that low BMI may reflect low muscle and high BMI fat mass (FM). BIA-derived lean and FM is directly associated with morbidity and mortality. To the contrary, BMI is rather of limited use for measuring BF% in epidemiological studies.

Obesity: lessons from evolution and the environment.

The 9th Stock Conference acknowledged the complex background of genetic, cultural, environmental and evolutionary factors of obesity. Gene-environment interactions underlie the flexibility in body-weight and body-fat regulation, illustrated by the hunter-gatherers' feast and famine lifestyle, the variation in physical activity over the lifespan being highest at reproductive age, the variation in energy intake through 'eating in the absence of hunger', while running the risk of exceeding the capacity of triacylglyceride storage, leading to lipotoxicity and metabolic problems. Perinatal metabolic programming for obesity via epigenetic changes in response to a 'Western diet' results in production of lipid-poor milk and metabolically efficient pups, contributing to the perpetuation of obesity throughout generations. Evolutionary insight from comparative physiology and ecology indicates that over generations activity-induced energy expenditure has remained the same compared to wild mammals, that energy balance might be dependant on protein balance, while the function of taste changed from detection of poison or energy to social drinking and social behaviour. At present, the impact of assortative mating on obesity prevalence is unambiguously positive. The complexity that appeared can only be fully appreciated by setting the data into the context of our evolutionary history.