Hypothesized confounders

Recently, the results of three prospective cohort mortality studies have been reported (44,82,83). The design of these cohort studies has addressed weaknesses in the cross-sectional analyses of population statistics. Cohort studies analyze the incidence of health effects in a sample of individuals whose relevant personal characteristics are recorded along with the exposures in question. Measurement of smoking, sex, age, occupation, and other individual characteristics for each participant in the study allow for direct adjustment for hypothesized confounders. 

Cross- sectional Design that simultaneously ascertains exposures and outcomes Appropriate for exploring research questions as a first attempt or to generate hypotheses for analytic study designs Can be completed in a relatively short period of time, often at reduced cost Useful initial approach for evaluating human health risks Disease prevalence only Cannot assess causality or the temporal relation between exposure and health outcome Residual confounding may threaten internal validity... 

The mechanisms for these interactions are not well defined, although numerous hypotheses have been advanced. It should be noted at the outset that attempts to determine the relationship between depression and altered immune function are confounded by the multiplicity of factors known to be associated with both and that may alter the interrelationship. Examples include age, gender, sleep status, the likelihood that depression represents a complex of disease states with varying involvement of the immune system, and the frequent presence of other psychiatric or physical disorders which can affect both psychological status and immune function. A detailed review of this topic has been published (hwin, 2001). Indeed, it has been pointed out by Irwdn that current data suggest immune changes in MDD specifically correlated with the disorder are also seen with stress and other psychiatric disorders, suggesting some common characteristic(s) shared by these problems (hwin, 2001). 

Interpretation of the effects evoked by the explanatory variables on injury or mortality involves hypotheses about injury causation as well as the expected trend. Since impact speed is a dominant determinant for injury and fatality, any explanatory variable that is associated with impact speed could act as a surrogate for impact speed and thus as a confounder, i.e., be significant without having a causal relationship or mask the original effect size due to the association with impact speed. Potential associations were tested using Pearson and Spearman correlations for continuous variables and t-tests and Mann-Whitney-Tests for non-continuous (binary) variables. P-values refer to the hypotheses of a correlation (for continuous variables) or to differences between the two groups (for binary variables). 

The above example highlights a fundamental difficulty with the correct interpretation of statistical results from observational (i.e., non-randomized) data sets. Since selection effects and dependencies between explanatory variables are common phenomena, a sound interpretation should include careful analysis of possible confounders and requires knowledge or hypotheses about cause-effect relations in advance based on different methods or sources of knowledge. 

To test the five hypotheses without the potential confounding effects of variation in the variables, we use two multi-item scales of measurements. One is 7-likert scale measuring system. The other is 5 point measuring system with specific questions that are easy to cause variation. For example, our mid-level managers inspect the work floor on a regular basis to check all operational processes. Different companies can interpret the regular check differently. Therefore we developed five items... 

Abstract— Pressure ulcer is primarily a biomechanical issue, although the cascade of clinical events often involves many confounding intrinsic and extrinsic factors, such as the general health condition and the personal hygiene of the subject involved. Decades of research have clarified some mechanisms leading to pressure ulcers, though considerable controversies still remain. A number of hypotheses have been proposed for the biomechanics of pressure ulcer in past decades - a) direct... 

It has been hypothesized that any lead-IQ association may not apply uniformly to all young children, but may be more pronounced in certain social groups. It is difficult to explore such statistical interactions between lead and confounding variables because of the large number of ways of classifying children into social subgroups. Nevertheless, some tentative observations may be of value. 

Cross-sectional studies are research methods that involve observing all of a population or a representative subset at a specific point in time. They collect data on outcomes and/or exposures collected on each participant at one moment in time. Thus, although they are simple and quick to perform, they are more robust at measuring associations with chronic diseases because they measure prevalent rather than incident outcomes. Cross-sectional studies that collect data on both outcome and exposure are not very robust in establishing the causal effect of an intervention, as they are prone to bias from confounding factors, but they can be used to test hypotheses about interventions and to justify a research objective. 

---