What Is the Association Between Depression and Cardiovascular Disease?

To the Editor We read with interest this commendable iteration of the Prospective Urban Rural Epidemiology (PURE) study by Rajan et al¹ describing an important association between depressive symptoms as assessed by the Short-Form Composite International Diagnostic Interview (CIDI-SF) and incident cardiovascular disease (CVD) and all-cause mortality.

Although this is the largest data set to determine this association, to our knowledge, its size does not limit bias resulting from selection, confounding, or reverse direction of the association reported. First, because of the numerous comparisons made from this single cohort, there are concerns of multiplicity and transparency. This is important because the use of the CIDI-SF questionnaire was not previously stated in the study protocol.² As previous studies used different thresholds (number of symptoms) to define major depression, it is paramount that analysis of this data are prespecified to prevent this effect.¹ Furthermore, even at a threshold of 5 or more symptoms, the CIDI-SF has a positive predictive value of 0.75, with individuals with false positives being older and less educated.³ These factors directly influence CVD and mortality. At baseline, while 11% of the patients fulfilled criteria of major depression, only 0.6% of them were taking antidepressants. Hence, we question the validity of CIDI-SF as the primary tool for depression assessment.