Long-term psychological stress and heart disease.

Sir—We were made aware of the interesting research conducted by Sibai et al. during our literature review. In their casecontrol study they investigated the relation of war stressors to coronary artery disease (CAD) as determined by coronary angiography and in their cohort study war exposures were studied with reference to CVD mortality. We find it difficult to compare their research with ours due to the different nature of the stressors studied. Lebanon is a country that emerged in 1991 from 16 years of conflict including multiple invasions and civil disturbances. It is interesting to hypothesize that the extreme level of stress over long periods of time experienced by their study population compared to the relatively variable sources and degrees of stress experienced by the subjects of our study population could explain the difference in results. It may be that only a high level of stress experienced over a long period of time can affect cardiovascular health. Authors found CAD to be related to the number of war events and frequency of crossing green lines. It is interesting to note that the OR associated with one war-related event was below one and an increased odds ratio was only observed after two or more events. As authors have acknowledged in their article concerning the case-control study, the risk of referral bias, misclassification of exposure, over-reporting and recall bias may at least in part explain their results. As noted by the authors, the heterogeneity of studies in this field does not allow for valid comparison. In order to correctly evaluate the differing aspects of psychological stress, a validated instrument should be used to measure exposure several times during follow-up in a large cohort. The use of regression models with time-dependant covariates could be a very useful statistical tool for analysing such data. © International Epidemiological Association 2000 Printed in Great Britain