Abstract
Background: Anxiety is not routinely assessed in patients hospitalized for acute myocardial infarction (AMI). Failure to identify and treat patients who are anxious after AMI makes them more vulnerable to the adverse effects of anxiety, including higher complication rates. The anxiety subscale of the Brief Symptom Inventory (BSI) is a simple, reliable measure of anxiety with minimal patient burden. However, there is limited evidence of reliability and validity of the BSI as a measure of anxiety in patients hospitalized for AMI.
Objective: The aim of this study was to provide evidence for the reliability and validity of the BSI in hospitalized AMI patients.
Method: A total of 536 patients admitted for AMI (62 +/- 14 years of age, 66% men, 85% white, 27% with previous myocardial infarction) completed the BSI and the state portion of the State Anxiety Inventory (SAI) within 72 hours of admission. Internal consistency reliability, criterion-related validity, and construct validity of the BSI were tested.
Results: There was sufficient evidence of internal consistency (Cronbach [alpha] = .87), which supported the reliability of the BSI. There was evidence of criterion-related validity based on the Spearman [rho] correlation coefficient of 0.70 (P < .001) between BSI and SAI scores. Anxious patients had higher complication rates than did nonanxious patients (BSI, 1.31 +/- 0.13 vs 0.82 +/- 0.08, respectively, P < .001; SAI, 1.34 +/- 0.13 vs 0.80 +/- 0.07, respectively, P < .001), demonstrating adequate construct validity. In a logistic regression, BSI scores were independent predictors of in-hospital complications after controlling for sociodemographic and clinical variables, which further supported construct validity.
Conclusion: These results provide support for the BSI as a reliable, valid instrument for measuring anxiety in patients hospitalized for AMI.