There is currently no consensus on the definition of weight maintenance in adults. Issues to consider in setting a standard definition include expert opinion, precedents set in previous studies, public health and clinical applications, comparability across body sizes, measurement error, normal weight fluctuations and biologic relevance. To be useful, this definition should indicate an amount of change less than is clinically relevant, but more than expected from measurement error or fluctuations in fluid balance under normal conditions. It is an advantage for the definition to be graded by body size and to be easily understood by the public as well as scientists. Taking all these factors into consideration, the authors recommend that long-term weight maintenance in adults be defined as a weight change of o3% of body weight.

The most popular measure for conducting analyses in studies of adiposity is body mass index (BMI); however, BMI does not discriminate between muscle and adipose tissue and does not directly assess regional adiposity. In this article, we address the question of whether alternatives to BMI should be used in epidemiologic analyses of the consequences of obesity. In general, measures of fat distribution such as waist circumference and sagittal abdominal diameter are more highly correlated with cardiovascular disease risk factors and diabetes than BMI; however, differences are usually small. Precise measures of adiposity from methods such as dual-energy x-ray absorptiometry may provide more specific and larger associations with disease, but published studies show that this is not always true. Further, practical considerations such as cost and feasibility must influence the choice of measure in many studies of large populations. Measures of adiposity are highly correlated with each other, and the additional cost of a more precise measure may not be justified in many circumstances. Validated prediction equations that include multiple anthropometric measures, along with demographic variables, may offer a practical means of obtaining assessments of total adiposity in large populations, whereas waist circumference can provide a feasible assessment of abdominal adiposity. Finally, public health messages to the public must be simple to be effective. Therefore, investigators may need to consider the ease of translation of results to the public when choosing a measure.

We examined the relationship of pericardial adipose tissue (PAT) with coronary artery calcification in MESA, a large cohort in which associations by race/ethnicity can be compared. The baseline cohort comprised 6,814 Caucasian (38%), African American (28%), Chinese American (12%) and Hispanic (22%) adults aged 45–84, without known clinical cardiovascular disease. Cardiac CT was used to measure PAT (cm3) and calcification (Agatston score). We examined cross-sectional associations of PAT with the presence (score>0) and severity (continuous score if >0) of calcification using prevalence ratio (PR) (n=6,672) and linear regression (n=3,362), respectively. Main models were adjusted for age, age2, gender, race/ethnicity, field site, smoking, physical activity, alcohol and education. PAT volume (adjusted for age, height, weight and site) was greatest in Chinese males, while Black males had less PAT than all but Black females. PAT was associated with presence [PR per standard deviation (SD): 1.06 (95% CI: 1.04, 1.08)] and severity [difference in log Agatston score per SD: 0.15 (0.09, 0.21)] of calcification, but neither association varied by race/ethnicity. Adjustment for generalized adiposity attenuated but did not eliminate the associations. With further adjustment for traditional risk factors and inflammatory markers, only the association with severity remained statistically significant [PR: 1.02 (1.00, 1.04), difference: 0.10 (0.03, 0.17)]. Heterogeneity by sex was observed for presence of calcification (PR in men: 1.04; in women: 1.08; p for interaction<0.0001). Pericardial adipose tissue was associated with the presence and severity of coronary artery calcification in this cohort, but despite differences in PAT volumes and calcification across race/ethnic groups, neither association varied by race/ethnicity.

No abstract

The choice of effect measure can influence the interpretation of data, particularly when comparing groups with substantially different event rates in the reference group. Since normal weight smokers have higher mortality than never smokers, we hypothesized that rate ratios (RRs) for the effect of BMI on mortality would be greater in never smokers than in smokers, but that rate differences (RDs) would be similar in both groups. Current and never smokers from the Atherosclerosis Risk in Communities Study (N=10,394) were aged 45–64 at baseline (1987–9) and followed through 2002. Poisson regression was used to estimate incidence rates for all‐cause mortality by BMI groups – normal weight (18.5–<25), overweight (25–<30) and obese (≥30). Models were adjusted for age, race, field center, education, physical activity, alcohol and cigarette‐years. Wald tests were used to determine if the RRs and RDs differed by smoking status. Mortality rates were higher in smokers than in never smokers (16.1 vs. 4.0/1000 person‐years in normal weight subjects). Among never smokers, rates were higher in the overweight and obese groups than in the normal weight group, but in smokers, they were lower in the overweight group and no different in the obese group. As hypothesized, the RRs for the effect of BMI on mortality were greater in never smokers than in smokers (p<.05) (Fig. ). However, the RDs were greater, as well (p<.05) (Fig. ). The attenuating effect of smoking on the BMI‐mortality relationship was not an artifact of the use of a ratio estimate. Smoking has potent effects on both BMI and mortality and therefore must be considered carefully in examinations of the BMI‐mortality relationship.