Resolving questions about bias in real and hypothetical referenda

Despite 10 years of research on behavior in hypothetical referenda, conflict remains in the literature on whether or not the mechanism generates biased responses compared to real referenda, and the nature and source of any such bias. Almost all previous inquiry in respect of this issue has concentrated on bias at the aggregate level. This paper reports a series of three experiments which focuses on bias at the individual level and how this can translate to bias at the aggregate level. The authors argue that only an individual approach to hypothetical bias is consistent with the concept of incentive compatibility. The results of these experiments reflect these previous conflicting findings but go on to show that individual hypothetical bias is a robust result driven by the differing influence of pure self-interest and other-regarding preferences in real and hypothetical situations, rather than by a single behavioral theory such as free riding. In a hypothetical situation these preferences cause yea-saying and non-demand revealing voting. This suggests that investigation of individual respondents in other hypothetical one-shot binary choices may also provide us with insights into aggregate behavior in these situations.

How community ecology links natural mortality, growth, and production of fish populations

Size-spectrum theory is used to show that (i) predation mortality is a decreasing function of individual size and proportional to the consumption rate of predators; (ii) adult natural mortality M is proportional to the von Bertalanffy growth constant K; and (iii) productivity rate P/B is proportional to the asymptotic weight W8 -1/3. The constants of proportionality are specified using individual level parameters related to physiology or prey encounter. The derivations demonstrate how traditional fisheries theory can be connected to community ecology. Implications for the use of models for ecosystem-based fisheries management are discussed.

Work-place social capital and smoking cessation: The Finnish Public Sector Study

AIMS:
To examine whether high social capital at work is associated with an increased likelihood of smoking cessation in baseline smokers.
DESIGN:
Prospective cohort study.
SETTING:
Finland.
PARTICIPANTS:
A total of 4853 employees who reported to be smokers in the baseline survey in 2000-2002 (response rate 68%) and responded to a follow-up survey on smoking status in 2004-2005 (response rate 77%).
MEASUREMENTS:
Work-place social capital was assessed using a validated and psychometrically tested eight-item measure. Control variables included sex, age, socio-economic position, marital status, place of work, heavy drinking, physical activity, body mass index and physician-diagnosed depression.
FINDINGS:
In multi-level logistic regression models adjusted for all the covariates, the odds for being a non-smoker at follow-up were 1.26 [95% confidence interval (CI)=1.03-1.55] times higher for baseline smokers who reported high individual-level social capital than for their counterparts with low social capital. In an analysis stratified by socio-economic position, a significant association between individual-level social capital and smoking cessation was observed in the high socio-economic group [odds ratio (OR) (95% CI)=1.63 (1.01-2.63)], but not in intermediate [(OR=1.10 (0.83-1.47)] or low socio-economic groups [(OR=1.28 (0.86-1.91)]. Work unit-level social capital was not associated with smoking cessation.
CONCLUSIONS:
If the observed associations are causal, these findings suggest that high perceived social capital at work may facilitate smoking cessation among smokers in higher-status jobs.

Social capital at work as a predictor of employee health: Multilevel evidence from work units in Finland

The majority of previous research on social capital and health is limited to social capital in residential neighborhoods and communities. Using data from the Finnish 10-Town study we examined social capital at work as a predictor of health in a cohort of 9524 initially healthy local government employees in 1522 work units, who did not change their work unit between 2000 and 2004 and responded to surveys measuring social capital at work and health at both time-points. We used a validated tool to measure social capital with perceptions at the individual level and with co-workers' responses at the work unit level. According to multilevel modeling, a contextual effect of work unit social capital on self-rated health was not accounted for by the individual's socio-demographic characteristics or lifestyle. The odds for health impairment were 1.27 times higher for employees who constantly worked in units with low social capital than for those with constantly high work unit social capital. Corresponding odds ratios for low and declining individual-level social capital varied between 1.56 and 1.78. Increasing levels of individual social capital were associated with sustained good health. In conclusion, this longitudinal multilevel study provides support for the hypothesis that exposure to low social capital at work may be detrimental to the health of employees. (c) 2007 Elsevier Ltd. All rights reserved.

Relationship between work stress and body mass index among 45,810 female and male employees

Objective: The proportion of overweight and obese people has grown rapidly, and obesity has now been widely recognized as an important public health problem. At the came time, stress has increased in working life. The 2 problems could be connected if work stress promotes unhealthy eating habits and sedentary behavior and thereby contributes to weight gain. This study explored the association between work stress and body mass index (BMI; kg/m(2)). Methods: We used cross-sectional questionnaire data obtained from 45,810 female and male employees participating in the ongoing Finnish Public Sector Cohort Study. We constructed individual-level scores, as well as occupational- and organizational-level aggregated scores for work stress, as indicated by the demand/control model and the effort-reward imbalance model. Linear regression analyses were stratified by sex and socioeconomic status (SES) and adjusted for age, marital status, job contract, smoking, alcohol consumption, physical activity, and negative affectivity. Results: The results with the aggregated scores showed that lower job control, higher job strain, and higher effort-reward imbalance were associated with a higher BMI. In men, lower job demands were also associated with a higher BMI. These associations were not accounted for by SES, although an additional adjustment for SES attenuated the associations. The results obtained with the individual-level scores were in the same direction, but the relationships were weaker than those obtained with the aggregated scores. Conclusions: This study shows a weak association between work stress and BMI.

Resident Aliens? Explaining Minority Disaffection with Democratic Politics in the Baltic States

Commentators and scholars alike recognize the important role political dissatisfaction plays in the process of regime change. A considerable body of literature has used dissatisfaction with a regime and distrust in political institutions to explain political dynamics during democratization's initial phase, yet these indicators are rarely used to assess disaffection with politics in established democratic regimes. Recent research on the post-communist region has established that citizens demonstrate high levels of political alienation, and that ethnic minority communities in particular are widely dissatisfied with democratic politics, institutions and regimes. This paper uses the 2004 data from the New Baltic Barometer to analyse individual-level disaffection with politics among the minorities in the Baltic States and explores the structural roots of such disaffection. The paper draws upon interviews with political representatives of minority communities in order to understand their perceptions of opportunities to participate in decision-making. Building on quantitative and qualitative analysis, the paper concludes that disaffection with politics among both the mass of ethnic minorities and their elite groups is best explained by the misrepresentation of minority interests in post-communist Baltic polities.

Fumonisin B-1 as a Urinary Biomarker of Exposure in a Maize Intervention Study Among South African Subsistence Farmers

Background: The consumption of maize highly contaminated with carcinogenic fumonisins has been linked to high oesophageal cancer rates. The aim of this study was to validate a urinary fumonisin B-1 (UFB1) biomarker as a measure of fumonisin exposure and to investigate the reduction in exposure following a simple and culturally acceptable intervention.

Methods: At baseline home-grown maize, maize-based porridge, and first-void urine samples were collected from female participants (n = 22), following their traditional food practices in Centane, South Africa. During intervention the participants were trained to recognize and remove visibly infected kernels, and to wash the remaining kernels. Participants consumed the porridge prepared from the sorted and washed maize on each day of the two-day intervention. Porridge, maize, and urine samples were collected for FB1 analyses.

Results: The geometric mean (95% confidence interval) for FB1 exposure based on porridge (dry weight) consumption at baseline and following intervention was 4.84 (2.87-8.14) and 1.87 (1.40-2.51) mg FB1/kg body weight/day, respectively, (62% reduction, P < 0.05). UFB1C, UFB1 normalized for creatinine, was reduced from 470 (295-750) at baseline to 279 (202-386) pg/mg creatinine following intervention (41% reduction, P = 0.06). The UFB1C biomarker was positively correlated with FB1 intake at the individual level (r - 0.4972, P < 0.01). Urinary excretion of FB1 was estimated to be 0.075% (0.054%-0.104%) of the FB1 intake.

Conclusion: UFB1 reflects individual FB1 exposure and thus represents a valuable biomarker for future fumonisin risk assessment.

Impact: The simple intervention method, hand sorting and washing, could positively impact on food safety and health in communities exposed to fumonisins. Cancer Epidemiol Biomarkers Prev; 20(3); 483-9. (C)2011 AACR.

Study of aflatoxin exposure in a Brazilian population using an aflatoxin - albumin biomarker

The evaluation of exposure to aflatoxins (AF) by measurement of the level of contamination in food is hampered due to the heterogeneous distribution of AF in food. Therefore, an alternative is to estimate the exposure using specific biological markers (biomarkers) based on an understanding of the metabolism of the compound. For AF, these include aflatoxin-N-7-guanine in the urine, or AFB(1)-albumin (AF-alb) in the blood. This study assessed the level of exposure to AF in Brazilian individuals using a biomarker approach, i.e. the AF-alb adducts. Blood samples were collected from urban residents (n=50; aged 18-52) in June 1999, at the Blood Center of Antonio Carlos de Camargo Hospital, Sao Paulo, Brazil. AF-alb adduct levels were determined, by ELISA following serum albumin extraction and digestion. AF-alb adducts were detected in 31/50 (62%) samples [range 0-57.3 pg AFB(1)-lys adducts/mg of blood albumin (pg/mg)]. The mean level of positives was 14.9 pg/mg and males had the two highest levels measured (57.1 and 57.3 pg/mg). There was no correlation with age or profession. This is the first study of Brazilian, or indeed South American, individuals that has determined exposure to AF at the individual level using a biomarker approach. These levels are similar to those observed in the Philippines. These data warrant further investigation of both the sources and consequences of exposure to this potent toxin in Brazil.

Determining mycotoxins and mycotoxigenic fungi in food and feed:Developing biomarkers of human exposure to mycotoxins

Exposure assessment is a critical part of epidemiological studies into the effect of mycotoxins on human health. Whilst exposure assessment can be made by estimating the quantity of ingested toxins from food analysis and questionnaire data, the use of biological markers (biomarkers) of exposure can provide a more accurate measure of individual level of exposure in reflecting the internal dose. Biomarkers of exposure can include the excreted toxin or its metabolites, as well as the products of interaction between the toxin and macromolecules such as protein and DNA. Samples in which biomarkers may be analysed include urine, blood, other body fluids and tissues, with urine and blood being the most accessible for human studies. Here we describe the development of biomarkers of exposure for the assessment of three important mycotoxins; aflatoxin, fumonisin and deoxynivalenol. A number of different biomarkers and methods have been developed that can be applied to human population studies, and these approaches are reviewed in the context of their application to molecular epidemiology research.

Modelling trajectories through the educational system in North West England

The main aim of this paper is to identify those school-level and locality-level factors that significantly affect each of the three stages in a young adult's educational trajectory in North West England: GCSE results, track taken at age 16 and 'A'-level scores. By applying three-level models to data collected as part of the EFFNATIS project, we find no evidence of any locality-level effects. Overall, none of the explanatory variables conventionally considered to affect educational attainment had a consistent effect across all three stages. Rather, each explanatory variable had a contingent effect at specific points within the overall trajectory of educational outcomes.

Proximity to a tobacco store and smoking cessation:A cohort study

Background
It is not clear whether the availability of tobacco affects the likelihood of smoking cessation. We examined whether the proximity to a tobacco store and
the number of stores were associated with smoking cessation, and compared results for proximity variables based on walking and straight-line (as the crow flies) distance.

Methods
The study population consisted of 8751 baseline smokers from the Finnish Public Sector study in 1997–2005. Smoking intensity (cigarettes/day) was
determined at baseline and smoking cessation was determined from a follow-up survey in 2008–2009. Proximity was measured using straight-line and walking
distance from home to the nearest tobacco store, and another exposure variable was the number of stores within 0.50 km from home. We calculated associations
with log-binomial regression models, adjusting for individual-level and area-level confounders.

Results
Of the participants, 3482 (39.8%) quit smoking during the follow-up (mean follow-up 5.5 years, SD 2.3 years). Among men who were moderate/heavy smokers at baseline and lived <0.50 km walking distance from the nearest tobacco store, the likelihood of smoking cessation was 27% (95% CI 12% to 40%) lower compared with those living ≥0.50 km from a store. Having even one store within 0.50 km walking distance from home decreased cessation in men who were moderate/heavy smokers by 37% (95% CI 19% to 51%). No decrease was found for men who were light smokers at baseline or for women.

Conclusions
Living within walking distance of a tobacco store reduced the likelihood of smoking cessation among men who were moderate/heavy smokers.

Hospitalization in Geriatric Wards: Considerations on the Evaluation of the Outcome

Within a defined law framework, the Italian central health system dictates the standards for hospitalization to local care units, which are in turn allowed to establish their own effectiveness criteria. The appropriateness of the hospitalization decision is therefore predetermined at patients admission, whereas its effectiveness relies on the ex post patient well-being as a result of the complex system of reciprocal relations between patients and healthcare agents at the ward level. We consider the outcomes in geriatric wards referring to the national health system, with respect both to patients traits at the individual level and wards/hospital settings. The risk that models the healthcare outcome is accordingly adjusted for covariates at the different levels of analysis (Goldstein & Spiegelhalter, 1996), thus allowing to differentiate among outcomes in terms of the hospitalization structure and, when appropriate, of territorial aggregation.

Genetically Determined Height and Coronary Artery Disease

Background: The nature and underlying mechanisms of an inverse association between adult height and the risk of coronary artery disease (CAD) are unclear. 
Methods: We used a genetic approach to investigate the association between height and CAD, using 180 height-associated genetic variants. We tested the association between a change in genetically determined height of 1 SD (6.5 cm) with the risk of CAD in 65,066 cases and 128,383 controls. Using individual-level genotype data from 18,249 persons, we also examined the risk of CAD associated with the presence of various numbers of height-associated alleles. To identify putative mechanisms, we analyzed whether genetically determined height was associated with known cardiovascular risk factors and performed a pathway analysis of the height-associated genes. 
Results: We observed a relative increase of 13.5% (95% confidence interval [CI], 5.4 to 22.1; P<0.001) in the risk of CAD per 1-SD decrease in genetically determined height. There was a graded relationship between the presence of an increased number of height-raising variants and a reduced risk of CAD (odds ratio for height quartile 4 versus quartile 1, 0.74; 95% CI, 0.68 to 0.84; P<0.001). Of the 12 risk factors that we studied, we observed significant associations only with levels of low-density lipoprotein cholesterol and triglycerides (accounting for approximately 30% of the association). We identified several overlapping pathways involving genes associated with both development and atherosclerosis. 
Conclusions: There is a primary association between a genetically determined shorter height and an increased risk of CAD, a link that is partly explained by the association between shorter height and an adverse lipid profile. Shared biologic processes that determine achieved height and the development of atherosclerosis may explain some of the association. (Funded by the British Heart Foundation and others.)

Relationships Between Deprivation and the Self-reported Health of Older People in Northern Ireland

Objectives: There are few studies on relationships between deprivation and the self-reported health of people aged over 64 years, and no studies fully representative of Northern Ireland’s older population. This paper addresses this gap. Methods: Deprivation of older people as reported in the 2001 and 2011 Censuses and the relationship with self-reported health are analyzed over a ten-year span using multilevel modeling. The data are from the Northern Ireland Longitudinal Study (NILS) linked to 2001-11 Census returns. Deprivation measures include housing tenure, property-value, access to a car, educational, employment and area-level income-deprivation. Results: Older people suffering deprivation face a significant health disadvantage over a ten-year time span. Discussion: This health disadvantage is stronger in men than in women, likely due to conservative gender roles prevalent among Northern Ireland’s older population, leading to psychological distress among deprived men. The analysis found strongly significant area-level effects, aggravating the health impact of deprivation.

People and places: Understanding geographical accuracy in administrative data from the census and healthcare systems

Administrative systems such as health care registration are of increasing importance in providing information for statistical, research, and policy purposes. There is thus a pressing need to understand better the detailed relationship between population characteristics as recorded in such systems and conventional censuses. This paper explores these issues using the unique Northern Ireland Longitudinal Study (NILS). It takes the 2001 Census enumeration as a benchmark and analyses the social, demographic and spatial patterns of mismatch with the health register at individual level. Descriptive comparison is followed by multivariate and multilevel analyses which show that approximately 25% of individuals are reported to be in different addresses and that age, rurality, education, and housing type are all important factors. This level of mismatch appears to be maintained over time, as earlier migrants who update their address details are replaced by others who have not yet done so. In some cases, apparent mismatches seem likely to reflect complex multi-address living arrangements rather than data error.