Epidemiological studies interpretation problems

Before considering the bioavailability of phytochemicals in a food matrix and relating these to biological effects in humans, it is important and necessary to understand that phytochemicals are mostly minor plant constituents whose concentration varies considerably according to, for example, seasonal and agronomic factors, the variety, age, and part of the plant examined. Such variability can lead to serious problems of interpretation of results from epidemiologal studies or human intervention trials if the dose of the phytochemical applied is not determined. 

All human studies dealing with chronic mustard disease processes are retrospective and fraught with the problems inherent in retrospective studies. These problems include bias in the sampling populations lack of epidemiological controls for the effects of smoking, lifestyle, race, gender, age, or exposure to other chemicals differential quality of available health care and incorrect diagnosis.6 These limitations make absolute interpretation of the studies difficult. 

It is not difficult to assume that the value of estimating cancer risk to man should be in the order epidemiologic studies > whole animal carcinogenesis assays > cell transformation and mutagenesis assays > chemical analysis alone. Each of these approaches has its own problems of interpretation and extrapolation, and requires many assumptions about how the human machine functions and lives. 

Epidemiological studies - methodology and interpretation. Problems in establishing past exposures of workers. Application of results to a particular workplace. Practicability of such studies in a single organisation. 

PROBLEMS IN INTERPRETING EXISTING DATA FROM EPIDEMIOLOGICAL STUDIES... 

Most data do not separate cement eczema into subgroups of allergic and irritant. Exact prevalences might therefore be difficult to give. Another problem in the interpretation of data concerning prevalences and incidences is due to the fact that most data are obtained from patch-test clinics. This may give inaccurate information due to selection bias. Studies carried out at work sites provide more accurate epidemiological information. 

---