Multiple imputation for body mass index: lessons from the Australian Longitudinal Study on Women's Health

In large epidemiological studies missing data can be a problem, especially if information is sought on a sensitive topic or when a composite measure is calculated from several variables each affected by missing values. Multiple imputation is the method of choice for 'filling in' missing data based on associations among variables. Using an example about body mass index from the Australian Longitudinal Study on Women's Health, we identify a subset of variables that are particularly useful for imputing values for the target variables. Then we illustrate two uses of multiple imputation. The first is to examine and correct for bias when data are not missing completely at random. The second is to impute missing values for an important covariate; in this case omission from the imputation process of variables to be used in the analysis may introduce bias. We conclude with several recommendations for handling issues of missing data. Copyright (C) 2004 John Wiley Sons, Ltd.

Effect of body mass index changes between ages 5 and 14 on blood pressure at age 14 - Findings from a birth cohort study

Weight reduction in clinical populations of severely obese children has been shown to have beneficial effects on blood pressure, but little is known about the effect of weight gain among children in the general population. This study compares the mean blood pressure at 14 years of age with the change in overweight status between ages 5 and 14. Information from 2794 children born in Brisbane, Australia, and who were followed up since birth and had body mass index (BMI) and blood pressure measurements at ages 5 and 14 were used. Systolic and diastolic blood pressure at age 14 was the main outcomes and different patterns of change in BMI from age 5 to 14 were the main exposure. Those who changed from being overweight at age 5 to having normal BMI at age 14 had similar mean blood pressures to those who had a normal BMI at both time points: age- and sex-adjusted mean difference in systolic blood pressure 1.54 ( - 0.38, 3.45) mm Hg and in diastolic blood pressure 0.43 ( - 0.95, 1.81) mm Hg. In contrast, those who were overweight at both ages or who had a normal BMI at age 5 and were overweight at age 14 had higher blood pressure at age 14 than those who had a normal BMI at both times. These effects were independent of a range of potential confounding factors. Our findings suggest that programs that successfully result in children changing from overweight to normal-BMI status for their age may have important beneficial effects on subsequent blood pressure.

An FTO variant is associated with Type 2 diabetes in South Asian populations after accounting for body mass index and waist circumference

Aims - A common variant, rs9939609, in the FTO (fat mass and obesity) gene is associated with adiposity in Europeans, explaining its relationship with diabetes. However, data are inconsistent in South Asians. Our aim was to investigate the association of the FTO rs9939609 variant with obesity, obesity-related traits and Type 2 diabetes in South Asian individuals, and to use meta-analyses to attempt to clarify to what extent BMI influences the association of FTO variants with diabetes in South Asians. Methods - We analysed rs9939609 in two studies of Pakistani individuals: 1666 adults aged = 40 years from the Karachi population-based Control of Blood Pressure and Risk Attenuation (COBRA) study and 2745 individuals of Punjabi ancestry who were part of a Type 2 diabetes case–control study (UK Asian Diabetes Study/Diabetes Genetics in Pakistan; UKADS/DGP). The main outcomes were BMI, waist circumference and diabetes. Regression analyses were performed to determine associations between FTO alleles and outcomes. Summary estimates were combined in a meta-analysis of 8091 South Asian individuals (3919 patients with Type 2 diabetes and 4172 control subjects), including those from two previous studies. Results - In the 4411 Pakistani individuals from this study, the age-, sex- and diabetes-adjusted association of FTO variant rs9939609 with BMI was 0.45 (95% CI 0.24–0.67) kg/m2 per A-allele (P = 3.0 × 10-5) and with waist circumference was 0.88 (95% CI 0.36–1.41) cm per A-allele (P = 0.001). The A-allele (30% frequency) was also significantly associated with Type 2 diabetes [per A-allele odds ratio (95% CI) 1.18 (1.07–1.30); P = 0.0009]. A meta-analysis of four South Asian studies with 8091 subjects showed that the FTO A-allele predisposes to Type 2 diabetes [1.22 (95% CI 1.14–1.31); P = 1.07 × 10-8] even after adjusting for BMI [1.18 (95% CI 1.10–1.27); P = 1.02 × 10-5] or waist circumference [1.18 (95% CI 1.10–1.27); P = 3.97 × 10-5]. Conclusions - The strong association between FTO genotype and BMI and waist circumference in South Asians is similar to that observed in Europeans. In contrast, the strong association of FTO genotype with diabetes is only partly accounted for by BMI.

The relationship of fear of intimacy, body image avoidance and body mass index

Issues of body image and ability to achieve intimacy are connected to body weight, yet remain largely unexplored and have not been evaluated by gender. The underlying purpose of this research was to determine if avoidant attitudes and perceptions of one's body may hold implications toward its use in intimate interactions, and if an above average body weight would tend to increase this avoidance. The National Health and Nutrition Examination Survey (NHANES, 1999-2002) finds that 64.5% of US adults are overweight, with 61.9% of women and 67.2% of men. The increasing prevalence of overweight and obesity in men and women shows no reverse trend, nor have prevention and treatment proven effective in the long term. The researcher gathered self-reported age, gender, height and weight data from 55 male and 58 female subjects (determined by a prospective power analysis with a desired medium effect size (r=.30) to determine body mass index (BMI), determining a mean age of 21.6 years and mean BMI of 25.6. Survey instruments consisted of two scales that are germane to the variables being examined. They were (1) Descutner and Thelen of the University of Missouri‘s (1991) Fear-of-Intimacy scale; and (2) Rosen, Srebnik, Saltzberg, and Wendt's (1991) Body Image Avoidance Questionnaire. Results indicated that as body mass index increases, fear of intimacy increases (p<0.05) and that as body mass index increases, body image avoidance increases (p<0.05). The relationship that as body image avoidance increases, fear of intimacy increases was not supported, but approached significance at (p<0.07). No differences in these relationships were determined between gender groups. For age, the only observed relationship was that of a difference between scores for age groups [18 to 22 (group 1) and ages 23 to 34 (group 2)] for the relationship of body image avoidance and fear of intimacy (p<0.02). The results suggest that the relationship of body image avoidance and fear of intimacy, as well as age, bear consideration toward the escalating prevalence of overweight and obesity. An integrative approach to body weight that addresses issues of body image and intimacy may prove effective in prevention and treatment.

Further evidence for the relationship between eating frequency, body mass index, and physical activity

The relationship between the frequency of eating, physical activity and Body Mass Index (BMI) was investigated. Seventy five women, aged 24 to 55, were recruited from Florida International University. Via interview, subjects provided information regarding demographics and habitual eating frequency over 24-hours, and completed both the Baecke Questionnaire of Habitual Physical Activity and the Health Insurance Plan of New York Questionnaire on Physical Activity. Pearson correlations and partial correlation coefficients were used to assess the relationship between eating frequency, physical activity, age, and BMI. Results revealed significant positive correlations between eating frequency and total physical activity scores, and leisure time physical activity scores, but not between eating frequency and physical activity on the job. Partial correlations suggest that there may be an effect of eating frequency on BMI both through an effect on physical activity and through another mechanism. These results suggest that more frequent eaters tend to be more physically active, which may partially explain why lower body weights is associated with more frequent eating.

The Relationship of Fear of Intimacy, Body Image Avoidance and Body Mass Index

Issues of body image and ability to achieve intimacy are connected to body weight, yet remain largely unexplored and have not been evaluated by gender. The underlying purpose of this research was to determine if avoidant attitudes and perceptions of one’s body may hold implications toward its use in intimate interactions, and if an above average body weight would tend to increase this avoidance. The National Health and Nutrition Examination Survey (NHANES, 1999-2002) finds that 64.5% of US adults are overweight, with 61.9% of women and 67.2% of men. The increasing prevalence of overweight and obesity in men and women shows no reverse trend, nor have prevention and treatment proven effective in the long term. The researcher gathered self-reported age, gender, height and weight data from 55 male and 58 female subjects (determined by a prospective power analysis with a desired medium effect size (r =.30) to determine body mass index (BMI), determining a mean age of 21.6 years and mean BMI of 25.6. Survey instruments consisted of two scales that are germane to the variables being examined. They were (1) Descutner and Thelen of the University of Missouri’s (1991) Fear-of-Intimacy scale and (2) Rosen, Srebnik, Saltzberg, and Wendt’s (1991) Body Image Avoidance Questionnaire. Results indicated that as body mass index increases, fear of intimacy increases (p<0.05) and that as body mass index increases, body image avoidance increases (p<0.05). The relationship that as body image avoidance increases, fear of intimacy increases was not supported, but approached significance at (p<0.07). No differences in these relationships were determined between gender groups. For age, the only observed relationship was that of a difference between scores for age groups [18 to 22 (group 1) and ages 23 to 34 (group 2)] for the relationship of body image avoidance and fear of intimacy (p<0.02). The results suggest that the relationship of body image avoidance and fear of intimacy, as well as age, bear consideration toward the escalating prevalence of overweight and obesity. An integrative approach to body weight that addresses issues of body image and intimacy may prove effective in prevention and treatment.

Evidence for three genetic loci involved in both anorexia nervosa risk and variation of body mass index

The maintenance of normal body weight is disrupted in patients with anorexia nervosa (AN) for prolonged periods of time. Prior to the onset of AN, premorbid body mass index (BMI) spans the entire range from underweight to obese. After recovery, patients have reduced rates of overweight and obesity. As such, loci involved in body weight regulation may also be relevant for AN and vice versa. Our primary analysis comprised a cross-trait analysis of the 1000 single-nucleotide polymorphisms (SNPs) with the lowest P-values in a genome-wide association meta-analysis (GWAMA) of AN (GCAN) for evidence of association in the largest published GWAMA for BMI (GIANT). Subsequently we performed sex-stratified analyses for these 1000 SNPs. Functional ex vivo studies on four genes ensued. Lastly, a look-up of GWAMA-derived BMI-related loci was performed in the AN GWAMA. We detected significant associations (P-values <5 × 10(-5), Bonferroni-corrected P<0.05) for nine SNP alleles at three independent loci. Interestingly, all AN susceptibility alleles were consistently associated with increased BMI. None of the genes (chr. 10: CTBP2, chr. 19: CCNE1, chr. 2: CARF and NBEAL1; the latter is a region with high linkage disequilibrium) nearest to these SNPs has previously been associated with AN or obesity. Sex-stratified analyses revealed that the strongest BMI signal originated predominantly from females (chr. 10 rs1561589; Poverall: 2.47 × 10(-06)/Pfemales: 3.45 × 10(-07)/Pmales: 0.043). Functional ex vivo studies in mice revealed reduced hypothalamic expression of Ctbp2 and Nbeal1 after fasting. Hypothalamic expression of Ctbp2 was increased in diet-induced obese (DIO) mice as compared with age-matched lean controls. We observed no evidence for associations for the look-up of BMI-related loci in the AN GWAMA. A cross-trait analysis of AN and BMI loci revealed variants at three chromosomal loci with potential joint impact. The chromosome 10 locus is particularly promising given that the association with obesity was primarily driven by females. In addition, the detected altered hypothalamic expression patterns of Ctbp2 and Nbeal1 as a result of fasting and DIO implicate these genes in weight regulation.

Body Mass Index as a measure of obesity

This Q and A style briefing paper provides an overview of the use of body mass index (BMI) for the measurement of obesity. BMI is a person’s weight in kilograms divided by the square of their height in metres. It is one of the most commonly used ways of estimating whether a person is overweight and hence more likely to experience health problems than someone with a healthy weight. It is used to measure population prevalence of overweight and obesity. It is also a relatively easy, cheap and non-invasive method for establishing weight status. However, BMI is only a proxy for body fatness. Factors such as fitness, ethnic origin and puberty can alter the relation between BMI and body fatness and must be taken into consideration. Other measurements such as waist circumference and skin thickness can be collected to indicate a person’s weight status or body fatness. None of these is as widely used as BMI.

Changes in children’s body mass index between 2006/07 and 2013/14 National Child Measurement Programme

This report is the sixth in a series of annual reports which use National Child Measurement Programme (NCMP) data to examine changes in children’s body mass index (BMI) that have taken place since 2006/07. It explores trends in obesity, overweight, excess weight and underweight prevalence, as well as changes in mean BMI over time. Trends within different socioeconomic and ethnic groups are also examined to determine whether existing health inequalities are widening or becoming smaller.

Application of body mass index adjusted for fat mass (BMIfat) obtained by bioelectrical impedance in adults

Introduction: Body mass index (BMI) has been one of the methods most frequently used for diagnose obesity, but it isn't consider body composition. Objective: This study intends to apply one new adiposity index, the BMI adjusted for fat mass (BMIfat) developed by Mialich, et al. (2011), in a adult Brazilian sample. Methods: A cross-sectional study with 501 individuals of both genders (366 women, 135 men) aged 17 to 38 years and mean age was 20.4 ± 2.8 years, mean weight 63.0 ± 13.5 kg, mean height 166.9 ± 9.0 cm, and BMI 22.4 ± 3.4 kg/m². Results and discussion: High and satisfactory R2 values were obtained, i.e., 91.1%, 91.9% and 88.8% for the sample as a whole and for men and women, respectively. Considering this BMIfat were developed new ranges, as follows: 1.35 to 1.65 (nutritional risk for malnutrition), > 1.65 and ≤ 2.0 (normal weight) and > 2.0 (obesity). The BMIfat had a more accurate capacity of detecting obese individuals (0.980. 0.993, 0.974) considering the sample as a whole and women and men, respectively, compared to the traditional BMI (0.932, 0.956, 0.95). Were also defined new cut-off points for the traditional BMI for the classification of obesity, i.e.: 25.24 kg/m² and 28.38 kg/m² for men and women, respectively. Conclusion: The BMIfat was applied for the present population and can be adopted in clinical practice. Further studies are needed to determine its application to different ethnic groups and to compare this index to others previously described in the scientific literature.

Factors Affecting the Body Mass Index in Adolescents in Portuguese Schools

The prevalence of overweight and obesity among children is increasing; hence, it was aimed to assess the Body Mass Index (BMI) in school children aged from 10 to 18 years for girls and to 17 for boys, as well as to identify the factors influencing BMI. This study included 742 students who answered a questionnaire previously approved for application in schools. The results showed some sociodemographic factors associated with BMI classes: age, school year, practicing high competition sport, being federate in a sport or having a vegetarian diet. The educational factors associated with BMI classes included only seminars given at school by a nutritionist. Behavioural factors significantly associated with BMI included: learning in classes, playing in the open air, reading books or use of internet. As conclusion, the results demonstrated that several factors affect BMI, and hence some actions could be taken in order to change them so as to reduce the prevalence of overweight, namely reinforcing the role of school and a more active participation of nutritionists in the education of the adolescents.

THE ASSOCIATION OF DEMOGRAPHIC VARIABLES AND DIETARY HABITS ON BODY MASS INDEX OF FOREIGN-BORN POPULATION IN SOUTH KOREA

The Republic of Korea (Korea) has been composed of a single ethnic group and, until very recently, has had little or no experience with large-volume of immigrants. The country, once known for its large-scale emigration, has emerged as a popular destination. As the foreign-born populations continue to grow, they experience dietary pattern changes. Dietary changes of immigrants are often associated with weight increase and consequently the risk of non-communicable diseases such as diabetes. This study examined the association of demographic variables and dietary habits on Body Mass Index (BMI) of foreign nationals living in Korea, based on a cross-sectional sample of 193 foreign-born men (120) and women (73). A web-based survey program, Zoho survey, was used to capture self-reported region of origin, socio-demographic variables, weight, height, frequency of physical activity, and food frequencies. For the entire sample (n=193), the World Health Organization (WHO) classification and the Korean Society for the Study of Obesity cut-off points of BMI were used. The average BMI was 24.6 (SD=3.6) kg/m2. Significant associations were observed between BMI and age (χ² (8, N=193) =26.22, p<.01), gender (χ² (2, N=193) =11.45, p<.01), and religious affiliation (χ² (4, N=191) =9.53, p<.05). On average, participants had meat and carbonated drinks 2.2 times a week; high calorie foods 1.5 times a week; and, dairy products 2.4 times a week. A significant correlation (r=.14, p<.05) between BMI and eating meat was also observed. This study showed that the BMI varied with age, gender and religion. Frequent consumption of meat was associated with a significant increase in BMI. This increase in BMI could be associated with the dietary changes experienced by the newcomers or absence of more familiar products. Consequently, as the volume of immigrants continues to increase in Korea, it is imperative to consider designing and implementing public health policies specifically in relation to a healthy diet that can impact the prevalence of preventable chronic diseases. Thus, health education among foreign-born population in Korea should focus on reducing the intake of meat and high energy foods.

The mediating role of dietary factors and leisure time physical activity on socioeconomic inequalities in body mass index among Australian adults

BACKGROUND: The relationship between socioeconomic position and obesity has been clearly established, however, the extent to which specific behavioural factors mediate this relationship is less clear. This study aimed to ascertain the contribution of specific dietary elements and leisure-time physical activity (LTPA) to variations in obesity with education in the baseline (1990-1994) Melbourne Collaborative Cohort Study (MCCS).

METHODS: 18, 489 women and 12, 141 men were included in this cross-sectional analysis. A series of linear regression models were used in accordance with the products of coefficients method to examine the mediating role of alcohol, soft drink (regular and diet), snacks (healthy and sweet), savoury items (healthy and unhealthy), meeting fruit and vegetable guidelines and LTPA on the relationship between education and body mass index (BMI).

RESULTS: Compared to those with lowest educational attainment, those with the highest educational attainment had a 1 kg/m2 lower BMI. Among men and women, 27% and 48%, respectively, of this disparity was attributable to differences in LTPA and diet. Unhealthy savoury item consumption and LTPA contributed most to the mediated effects for men and women. Alcohol and diet soft drink were additionally important mediators for women.

CONCLUSIONS: Diet and LTPA are potentially modifiable behavioural risk factors for the development of obesity that contribute substantially to inequalities in BMI. Our findings highlight the importance of specific behaviours which may be useful to the implementation of effective, targeted public policy to reduce socioeconomic inequalities in obesity.

Future predictions of body mass index and overweight prevalence in Australia, 2005-2025

To predict current and future body mass index (BMI) and prevalence of overweight and obesity in Australian children and adults based on sex, age and year of birth (cohort). These predictions are needed for population health planning and evaluation. Data were drawn from 11 cross-sectional national or state population surveys conducted in Australia between 1969 and 2004. These included representative population samples of children (n= 27,635) and adults (n= 43,447) aged 5 years or older with measured height and weight data. Multiple linear regression analyses of measured log-transformed BMI data were conducted to determine the independent effects of age and year of birth (cohort) on ln(BMI) for males and females, respectively. Regression coefficients for cohort obtained from these analyses were applied to the National Nutrition Survey 1995 data set to predict mean BMI and prevalence of overweight (BMI 25-29.99 kg/m(2)) and obesity (BMI ≥ 30 kg/m(2)) in 2005, 2015 and 2025. Based on past trends, BMI is predicted to continue to increase for both males and females and across the age span. This would result in increases in the prevalence of overweight and obesity of between 0.4 and 0.8% per year, such that by 2025 around one-third of 5-19 year olds will be overweight or obese as will 83% of males and 75% of females aged 20 years and over. The increases in prevalence and mean BMI predicted in this study will have significant impacts on disease burden, healthcare costs and need for prevention and treatment programmes.