Figure. Emergency department crowding in tertiles. PTSD indicates posttraumatic stress disorder. Error bars indicate 95% confidence intervals.
Edmondson D, Shimbo D, Ye S, Wyer P, Davidson KW. The association of emergency department crowding during treatment for acute coronary syndrome predicts with subsequent posttraumatic stress disorder symptoms. JAMA Internal Med. Published online February 11, 2013. doi:10.1001/jamainternmed.2013.2536
eTable 1. Participant characteristics by degree of emergency department crowding during treatment for acute coronary syndrome
Edmondson D, Shimbo D, Ye S, Wyer P, Davidson KW. The Association of Emergency Department Crowding During Treatment for Acute Coronary Syndrome With Subsequent Posttraumatic Stress Disorder Symptoms. JAMA Intern Med. 2013;173(6):472-475. doi:10.1001/jamainternmed.2013.2536
Author Affiliations: Center for Behavioral Cardiovascular Health, Departments of Medicine (Drs Edmondson, Shimbo, Ye, and Davidson) and Emergency Medicine (Dr Wyer), Columbia University Medical Center, New York, New York.
Our recent meta-analysis of 24 studies (N = 2383 patients) found that 12% of patients with acute coronary syndrome (ACS) (including patients with ST-segment elevation myocardial infarction [STEMI], non-STEMI, and unstable angina [UA]) develop posttraumatic stress disorder (PTSD) symptoms owing to their ACS, and that these PTSD symptoms 1 month post-ACS are associated with a doubling of risk for ACS recurrence or mortality in the subsequent 1 to 3 years.1 Although we reviewed the known risk factors for ACS-induced PTSD, including demographic and psychological factors, we noted that no study has addressed the possibility that the medical environments in which initial ACS diagnosis and treatment occur might contribute to the traumatic nature of patients' experience during their ACS events. The emergency department (ED) is the initial medical environment for most patients with ACS, and the ED they enter is often overcrowded and chaotic.2 This overcrowding problem has recently been identified as a public health problem by the Institute of Medicine,3 due in part to associated delays in treatment and lower patient satisfaction. Emergency department crowding may contribute to poor cardiovascular outcomes in patients who present with chest pain,4 and we believe that crowded EDs may also contribute to increased risk for ACS-induced PTSD. We hypothesized that greater ED crowding would be associated with increased ACS-induced PTSD symptoms 1 month after presentation at an ED with an ACS event.
We recorded the time of presentation to the ED of a large urban teaching hospital in New York City for 135 participants (eTable) in the ongoing Prescription Use, Lifestyle, Stress Evaluation (PULSE) observational cohort study between 2009 and 2011. These patients were selected because they met the study criteria for ACS and were admitted through the ED during the study period. We determined that mean length of stay (LOS) in the ED was more than 11 hours. We calculated the degree of ED crowding exposure for each participant by summing the hourly ED admissions for the 12 hours prior to and the 12 hours subsequent to each participant's ED presentation time, and we divided the score into tertiles for analysis. Similar metrics have been used previously, but no gold standard for ED crowding measurement currently exists.5 We assessed demographic characteristics by self-report and abstracted clinical data from medical charts. We assessed in-hospital depression symptoms with the Beck Depression Inventory (BDI), and ACS -induced PTSD symptoms by telephone interview 1 month later using the Impact of Events Scale-Revised6 (IES-R) specific for ACS. We tested for group differences across ED crowding tertiles using 1-way analysis of variance for continuous variables and χ2 for categorical variables. We used multiple linear regression to assess the association of ED crowding to PTSD symptoms at 1 month.
Participants did not differ on any covariate by ED crowding tertile. Their characteristics were as follows: the mean (SD) age, 63.3 (10.7); 72% were men; 23% were African American; and 47% were Hispanic. Thirteen percent had STEMI; 32%, non-STEMI; 55%, UA; 33%, prior myocardial infarction (MI). Their mean (SD) Global Registry of Acute Coronary Events (GRACE) score was 91.2 (29.1); the left ventricular ejection fraction was less than 40 in 16%; the mean (SD) Charlson comorbidity index score was 1.8 (1.7); and the mean BDI score was 8.4 (7.4). Increasing tertiles of ED crowding were associated with higher levels of 1-month ACS-induced PTSD symptoms in univariate analysis (B = 2.0; P < .05). After adjustment for age, sex, education, race/ethnicity, previous MI, GRACE mortality risk score, Charlson comorbidity index, left ventricular ejection fraction of 40% or higher, and in-hospital BDI depression score, increasing tertiles of ED crowding were associated with higher ACS-induced PTSD symptoms at 1 month (B = 2.5; P = .01) (Figure). Further adjustment for ED LOS did not alter the result (B = 3.0; P = .02; N = 99 patients with complete LOS data), and a sensitivity analysis restricted to the 118 patients with non-STEMI and UA yielded nearly identical results.
Our results suggest that ED crowding may be associated with ACS-induced PTSD symptoms, a risk factor for ACS recurrence and mortality, and a known contributor to poor quality of life, patient satisfaction, and increased medical utilization. Furthermore, the association is independent of ACS severity, medical comorbidities, demographic factors, and depression. To our knowledge, this is the first study to test the relationship of objective features of the ED to the development of subsequent PTSD symptoms in patients with ACS. A mechanism for the association of ED crowding to increased PTSD symptoms may be that a more chaotic environment may foster or inflate perceptions of increased life threat and decreased control, which may in turn contribute to greater acute psychological and physiological arousal.7 Furthermore, ED crowding may be associated with poor physician-patient communication, which has been shown to be related to subsequent PTSD in other patient populations.8
Although our results are based on a small sample from a single ED, we believe they suggest the need for greater awareness of the influence of medical environments on patients' psychological well-being, while underscoring the need for hospital administrators and policymakers to address ED overcrowding.
Correspondence: Dr Edmondson, Center for Behavioral Cardiovascular Health, Columbia University Medical Center, 622 W 168 St, New York, NY 10032 (firstname.lastname@example.org).
Published Online: February 11, 2013. doi:10.1001/jamainternmed.2013.2536
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by grant HL-088117 and in part by Columbia University's CTSA grant No. UL1RR024156 from the National Center for Advancing Translational Sciences-National Center for Research Resources (NCRR)/National Institutes of Health (NIH), Bethesda, Maryland. Dr Edmondson is supported by KM1 grant No. CA-156709. Dr Ye is supported by NIH grant T32HL007854-16.
Disclaimer: The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official view of the NCRR or the NIH.