Epidemiological Studies of the Association between Psychosocial Factors and Gastrointestinal Disorders

Abstract

Three different epidemiological studies of psychosocial factors and gastrointestinal illness are described. Problems of illness definition, representativeness of samples and methods are discussed. It is emphasized that a full understanding of the possible role of psychosocial factors in gastrointestinal illness requires studies not only of patients who consult physicians. Case finding efforts or studies of all (or random samples of all) subjects in a population are required.

Findings on a population of middle-aged building-construction workers indicate that the factors ‘having grown up as a late child in a large family (at least nr 7)″, ‘hostility (with slow persons and in queues)', ‘broke up from previous marriage’, and ‘currently a cigarette smoker’, all predicted increased risk of near-future long-lasting work absenteeism (at least 30 consecutive days) because of ulcer or gastritis. Two other studies indicated the importance of low decision latitude on the job and bad social support at work in increasing the risk of self-reported gastrointestinal illness.

In order to improve our understanding of the pathogenesis of gastrointestinal disorders, we must explore how these disorders arise in all kinds of people–not only those who happen to be our patients.

It is difficult to perform good epidemiological studies of psychosocial factors and gastrointestinal illness. Apart from the difficulties of finding relevant and reliable measures of psychosocial factors, it is also difficult to find cases of illness themselves. A large proportion of those who suffer from gastritis or irritable bowel syndrome never consult physicians, and therefore a search for cases in official patient registers may not provide the researcher with a representative group of suffering subjects. It may be necessary to send questionnaires to all(or a random group of all) subjects belonging to a population in order to ascertain true representativeness. However, in the latter case it is difficult to get access to information from large sample of subjects, which may be necessary if social factors are being explored. Furthermore, clinicians are never satisfied with questionnaire data alone since subjective symptoms are not always associated with physiological changes. The reverse is also true: Many physiological changes are not associated with subjective symptoms.

Because of all these difficulties, it may be necessary to combine information obtained from studies utilizing widely different methodologies and sampling procedures. I shall concentrate on studies which have used either case finding techniques–efforts to search for allcases of gastrointestinal disorder in a group or area comparing them with matched non-cases–or studies exploring whole groups or representative samples of defined populations. Furthermore, the studies that I chose to discuss have explored: