In Reply In our study examining the association of depressive disorders with incident cardiovascular disease in adults with human immunodeficiency virus (HIV) infection,1 we applied a conservative approach to exclude any participant who had 1 inpatient or outpatient International Classification of Diseases, Ninth Revision code at baseline for acute myocardial infarction (AMI), unstable angina, cardiovascular revascularization, stroke, or heart failure in any of our data sources (Veterans Affairs, Veterans Affairs fee for service, Medicare, and Medicaid files). Although 1 outpatient code would not qualify as an event from an outcomes perspective, eliminating such participants helped to minimize the possibility that prevalent cardiovascular disease was operating as a confounder. That said, Ziegelstein is correct that we could not eliminate prevalent subclinical atherosclerosis. As measures of subclinical atherosclerosis are not routinely collected in asymptomatic people, such data were not available, and thus, we cannot know whether prevalent subclinical atherosclerosis predisposed participants to a proinflammatory state that led to depressive disorder onset and incident AMI. However, our results are consistent with prior work from members of our team. In the Pittsburgh Healthy Heart Project,2 depressive symptoms were associated with future progression of subclinical atherosclerosis, even after adjusting for confounders and baseline subclinical atherosclerosis.
Freiberg MS, Stewart JC, Khambaty T. Association of Depression and Cardiovascular Disease—Reply. JAMA Cardiol. 2017;2(6):703. doi:10.1001/jamacardio.2016.5999