Coronary Procedures and Outcomes in Veterans With PTSD
Our findings should be considered in light of the following limitations. First, residual confounding is possible given the observational nature of our study, despite our inclusion of detailed clinical and administrative data. Second, our findings represent the experience of veterans with PTSD undergoing coronary procedures in the VA and may not generalize to other settings given that the VA represents an integrated, single-payer health care system that serves US veterans who are predominantly older, male, and with a high prevalence of PTSD and comorbid disease. Furthermore, the VA surveys patients with the Patient Health Questionnaire (PHQ) and the PHQ-9, screening questionnaires for mental health disorders that may contribute to identification of patients with these psychiatric disorders relative to the non-VA population. Third, the use of diagnosis codes for PTSD, comorbid depression, anxiety, and alcohol and substance use disorders may lead to patient misclassification. We sought to minimize misclassification by limiting diagnosis codes to clinical visits occurring within 1 year before the coronary procedure. However, use of clinical measures of PTSD may result in more refined assessments of the disease. For example, recent studies suggest that specific symptoms of PTSD, particularly intrusive symptoms, may be associated with increased risk for cardiovascular disease, whereas other PTSD symptoms are not. Thus, measuring specific PTSD symptoms or symptom clusters and understanding their relationship to cardiac outcomes may provide different findings. Fourth, we did not adjust for medication use for psychiatric diagnoses given challenges of indication bias that we cannot account for in the absence of detailed clinical data on PTSD severity. Finally, as we did not have detailed data on clinicians' rationale for pursuing coronary angiography, we could not determine the exact reasons for the apparent increased use of coronary angiography for elective indications in patients with a PTSD diagnosis.
Limitations
Our findings should be considered in light of the following limitations. First, residual confounding is possible given the observational nature of our study, despite our inclusion of detailed clinical and administrative data. Second, our findings represent the experience of veterans with PTSD undergoing coronary procedures in the VA and may not generalize to other settings given that the VA represents an integrated, single-payer health care system that serves US veterans who are predominantly older, male, and with a high prevalence of PTSD and comorbid disease. Furthermore, the VA surveys patients with the Patient Health Questionnaire (PHQ) and the PHQ-9, screening questionnaires for mental health disorders that may contribute to identification of patients with these psychiatric disorders relative to the non-VA population. Third, the use of diagnosis codes for PTSD, comorbid depression, anxiety, and alcohol and substance use disorders may lead to patient misclassification. We sought to minimize misclassification by limiting diagnosis codes to clinical visits occurring within 1 year before the coronary procedure. However, use of clinical measures of PTSD may result in more refined assessments of the disease. For example, recent studies suggest that specific symptoms of PTSD, particularly intrusive symptoms, may be associated with increased risk for cardiovascular disease, whereas other PTSD symptoms are not. Thus, measuring specific PTSD symptoms or symptom clusters and understanding their relationship to cardiac outcomes may provide different findings. Fourth, we did not adjust for medication use for psychiatric diagnoses given challenges of indication bias that we cannot account for in the absence of detailed clinical data on PTSD severity. Finally, as we did not have detailed data on clinicians' rationale for pursuing coronary angiography, we could not determine the exact reasons for the apparent increased use of coronary angiography for elective indications in patients with a PTSD diagnosis.
SHARE
