Statistical approaches to testing the relationships of the built environment with resident-level physical activity behavior and health outcomes in cross-sectional studies with cluster sampling

To achieve valid conclusions, studies exploring associations of the built environment with residents' physical activity and health-related outcomes need to employ statistical approaches accounting for clustered data. This article discusses the following main statistical approaches: analysis of covariance, regression models with robust standard errors, generalized estimating equations, and multilevel generalized linear models. The choice of a statistical method depends on the characteristics of the study and research questions. While the first three approaches are employed to account for clustering in the data, multilevel models can also help unravel more substantive issues within a social ecological theoretical framework of health behavior.

Investor heterogeneity and the cross-sectional stock returns in China

We examine the relationship between divergence of opinion and the cross-sectional stock returns in Chinese A share market where short-selling of stocks is prohibited by law. Using a proxy for divergence of opinion among the entire investor base, we document a positive relationship between divergent beliefs and future stock returns. This is in sharp contrast to Miller's (1977) prediction of a negative relationship between the two. The result is likely to be driven by the dominance of individual investors and their speculative trading behaviors in China. Miller's prediction is confirmed when divergence of opinion is measured using data on mutual fund holdings. Our results are robust to a number of common return predictors. We also find a significantly negative relationship between the fraction of tradable shares in listed Chinese companies and future stock returns. Increase in the fraction of tradable shares tends to reduce the predictability of stock returns using divergence of opinion.

Components of the diet associated with child adiposity: a cross-sectional study

Objective:
This study sought to determine which components of youths' diets were related to adiposity while controlling for potential often-neglected confounders such as moderate to vigorous physical activity (MVPA) and dietary reporting error. Secondary goals of this study were to determine the extent to which MVPA confounded the associations between diet and adiposity and whether associations between diet and adiposity would differ depending on reporting error.

Methods:
An ethnically diverse urban sample of 342 children aged 9–10 years and 323 adolescents aged 17–18 years were recruited for this cross-sectional study. Body mass index (BMI) and waist circumference (WC) were measured in the school; dietary assessment included three 24-hour recalls via telephone in the evenings, and MVPA assessment included 5 days of accelerometry. Over (n = 68), under (n = 250), or plausible (n = 347) dietary intake reporters were identified with the Huang calculation method. Linear regression assessed the relationship between adiposity indicators (BMI z-score and WC) and components of the diet (energy intake, food groups, macronutrients) after controlling for reporting error, demographic variables, and MVPA.

Results:
When dietary reporting error and potential confounders such as MVPA and demographic variables were controlled, energy intake (EI), vegetables, refined grains, total fat, total protein, and total carbohydrate were positively related to BMI z-score and WC and artificially sweetened beverages to WC. MVPA was a significant confounder. For BMI z-score, but not WC, relationships and strength of these relationships differed depending on dietary reporting error group (plausible, underreporter, overreporter).

Conclusions:
Among plausible reporters, as expected, EI, refined grains, and all macronutrients were positively related to adiposity; however, artificially sweetened beverages and vegetables, which are low-energy-dense foods, were also positively related to adiposity. Reporting error interfered with associations between diet and BMI z-score but not WC, suggesting WC is a more robust measure of adiposity in relation to diet.

Investigating well-being, work limitations and preferences for self-management education and peer support among younger people with hip and knee osteoarthritis: protocol for a cross-sectional study

Introduction: 

Relationship between overweight and health-related quality of life in secondary school children in Fiji: results from a cross-sectional population-based study

Objective
To investigate the relationship between excess weight (overweight and obesity) and health-related quality of life (HRQoL) in a sample of secondary school children in Fiji, by gender, age and ethnicity.

Methods
The study comprised 8947 children from forms 3–6 (age 12–18 years) in 18 secondary schools on Viti Levu, the main island of Fiji. Body mass index (BMI) was calculated from measured height and weight, and weight status was classified according to the International Obesity Task Force recommendations. HRQoL was measured by the self-report version of the Pediatric Quality of Life Inventory 4.0.

Results
HRQoL was similar in children with obesity and normal weight. Generally, this was replicated when analyzed separately by gender and ethnicity, but age stratification revealed disparities. In 12–14-year-old children, obesity was associated with better HRQoL, owing to better social and school functioning and well-being, and in 15–18-year olds with poorer HRQoL, owing to worse physical, emotional and social functioning and well-being (Cohen’s d 0.2–0.3). Children with a BMI in the overweight range also reported a slightly lower HRQoL than children with a BMI in the normal weight range, but although statistically significant, the size of this difference was trivial (Cohen’s d <0.2).

Discussion
The results suggest that, overall there is no meaningful negative association between excess weight and HRQoL in secondary school children in Fiji. This is in contradiction to the negative relationship between excess weight and HRQoL shown in studies from other countries and cultures. The assumption that a large body size is associated with a lower quality of life cannot be held universally. Although a generally low HRQoL among children in Fiji may be masking or overriding the potential effect of excess weight on HRQoL, socio-economic and/or socio-cultural factors, may help to explain these relationships.

Independent and joint associations of TV viewing time and snack food consumption with the metabolic syndrome and its components; a cross-sectional study in Australian adults

Background:
Television (TV) viewing time is positively associated with the metabolic syndrome (MetS) in adults. However, the mechanisms through which TV viewing time is associated with MetS risk remain unclear. There is evidence that the consumption of energy-dense, nutrient poor snack foods increases during TV viewing time among adults, suggesting that these behaviors may jointly contribute towards MetS risk. While the association between TV viewing time and the MetS has previously been shown to be independent of adult’s overall dietary intake, the specific influence of snack food consumption on the relationship is yet to be investigated. The purpose of this study was to examine the independent and joint associations of daily TV viewing time and snack food consumption with the MetS and its components in a sample of Australian adults.

Methods:
Population-based, cross-sectional study of 3,110 women and 2,572 men (>35 years) without diabetes or cardiovascular disease. Participants were recruited between May 1999 and Dec 2000 in the six states and the Northern Territory of Australia. Participants were categorised according to self-reported TV viewing time (low: 0-2 hr/d; high: >2 hr/d) and/or consumption of snack foods (low: 0-3 serves/d; high: >3 serves/d). Multivariate odds ratios [95% CI] for the MetS and its components were estimated using gender-specific, forced entry logistic regression.

Results:
OR [95% CI] for the MetS was 3.59 [2.25, 5.74] (p≤0.001) in women and 1.45 [1.02, 3.45] (p = 0.04) in men who jointly reported high TV viewing time and high snack food consumption. Obesity, insulin resistance and hypertension (women only) were also jointly associated with high TV viewing time and high snack food consumption. Further adjustment for diet quality and central adiposity maintained the associations in women. High snack food consumption was also shown to be independently associated with MetS risk [OR: 1.94 (95% CI: 1.45, 2.60), p < 0.001] and hypertension [OR: 1.43 (95% CI: 1.01, 2.02), p = 0.05] in women only. For both men and women, high TV viewing time was independently associated with the MetS and its individual components (except hypertension).

Conclusion:
TV viewing time and snack food consumption are independently and jointly associated with the MetS and its components, particularly in women. In addition to physical activity, population strategies targeting MetS prevention should address high TV time and excessive snack food intake.

Excessive alcohol use and its association with risky sexual behaviour: a cross-sectional analysis of data from Victorian secondary school students

Objective
Estimate the prevalence of sexual behaviour and alcohol use and examine the association between excessive alcohol use and risky sexual behaviour in late secondary students in Victoria, Australia.

Method
The sample of Year 11 students from government and independent schools participating in the 2008 International Youth Development Study (n=450) was representative of the Victorian school population. Logistic regression analyses examined the associations between sexual behaviour, binge and compulsive drinking, adjusting for socio-demographic, school and family factors.

Results
Under half (44%) the students had experienced sex in the past year, half (50%) had engaged in binge drinking in the past two weeks and 26% reported compulsive drinking in the past year. Of those who reported sex in the past year (n=197), 34% had sex without a condom at the last sexual encounter and 28% later regretted sex due to alcohol. The likelihood of experiencing sex was increased by binge (OR=2.44, 95%CI 1.44-4.12) and compulsive drinking (OR=2.15, 95%CI 1.29-3.60). For those sexually active, binge drinking increased the risk of having three or more sexual partners (OR=3.37, 95%CI 1.11-10.26) and compulsive drinking increased the likelihood of regretted sex due to alcohol (OR=4.43, 95%CI 2.10-9.31). Excessive drinking was not associated with condom non-use.

Conclusion and implications
Risky sex – multiple sexual partners and regretted sex due to alcohol – and excessive drinking are highly prevalent and co-associated among Victorian late secondary students.

Mid-femoral and mid-tibial muscle cross-sectional area as predictors of tibial bone strength in middle-aged and older men

While it is widely acknowledged that bones adapt to the site-specific prevalent loading environment, reasonable ways to estimate skeletal loads are not necessarily available. For long bone shafts, muscles acting to bend the bone may provide a more appropriate surrogate of the loading than muscles expected to cause compressive loads. Thus, the aim of this study was to investigate whether mid-thigh muscle cross-sectional area (CSA) was a better predictor of tibial mid-shaft bone strength than mid-tibia muscle CSA in middle aged and older men. 181 Caucasian men aged 50–79 years (mean±SD; 61±7 years) participated in this study. Mid-femoral and mid-tibial bone traits cortical area , density weighted polar moment of area and muscle CSA [cm²] were assessed with computed tomography. Tibial bone traits were positively associated with both the mid-femur (r=0.44 to 0.46, P<0.001) and the mid-tibia muscle CSA (r=0.35 to 0.37, P<0.001). Multivariate regression analysis, adjusting for age, weight, physical activity and femoral length, indicated that mid-femur muscle CSA predicted tibial mid-shaft bone strength indices better thn mid-tibia muscle CSA. In conclusion, the association between a given skeletal site and functionally adjacent muscles may provide a meaningful probe of the site-specific effect of loading on bone.

Does workplace social capital buffer the effects of job stress? A cross-sectional, multilevel analysis of cigarette smoking among U.S. manufacturing workers

Objective: To investigate whether workplace social capital buffers the association between job stress and smoking status. Methods: As part of the Harvard Cancer Prevention Project's Healthy Directions—Small Business Study, interviewer-administered questionnaires were completed by 1740 workers and 288 managers in 26 manufacturing firms (84% and 85% response). Social capital was assessed by multiple items measured at the individual level among workers and contextual level among managers. Job stress was operationalized by the demand-control model. Multilevel logistic regression was used to estimate associations between job stressors and smoking and test for effect modification by social capital measures. Results: Workplace social capital (both summary measures) buffered associations between high job demands and smoking. One compositional item—worker trust in managers—buffered associations between job strain and smoking. Conclusion: Workplace social capital may modify the effects of psychosocial working conditions on health behaviors.