Copyright 2006 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2006
I read with interest the analysis published recently by Giltay et al entitled “Dispositional Optimism and the Risk of Cardiovascular Death.”1 I was surprised, however, that the authors did not also report noncardiovascular and all-cause mortality in association with optimism. There may be a priori reasons to think that cardiovascular deaths are more tightly associated with optimism compared with other types of death, but given the uncertainty inherent in determining cause of death (as noted by the authors and others2), it seems prudent to analyze noncardiovascular and all-cause mortality at least as secondary outcomes. Perhaps optimistic patients were also “optimistic” about chest pain symptoms, did not report them to their physicians, and died suddenly without being diagnosed as having coronary disease at higher rates compared with less optimistic persons. This type of differential measurement error in outcome ascertainment could bias the findings in the direction observed. On the other hand, if all-cause mortality shows at least a trend in the same direction as the cardiovascular mortality analysis and noncardiovascular disease mortality shows no trend in the opposite direction, I would be reassured.
Pletcher MJ. What About Other Causes of Death?. Arch Intern Med. 2006;166(14):1528. doi:10.1001/archinte.166.14.1528-b