Inconsistent Reporting of Adverse Life Events Is Predicted by Current Internalizing Distress Among Military Service Members

Abstract

Introduction: Accurate measurement of adverse life events is critical for understanding the effects of stressors on health outcomes. However, much of this research uses cross-sectional designs and self-report years after the events take place. The reliability of this retrospective reporting and the individual difference factors associated with inconsistent recall over time are not frequently addressed, especially among military service members.

Materials and methods: A longitudinal cohort of National Guard service members (n = 801) completed the Deployment Risk and Resilience Inventory-2 Prior Stressors scale and several measures of general well-being, including anxious depressive symptomatology, personal functioning, perceived social support, and overall health at two time points (before and after completion of basic combat training; median 11-month interval).

Results: Consistency in reporting the life event items ranged from 69.5% to 99.7%, with an overall Cohen's kappa coefficient of 0.215 for the scale, indicating minimal agreement. Lower well-being scores at Time 1 independently predicted yes-to-no changes in responding, whereas lower well-being scores at Time 2 independently predicted no-to-yes changes in responding. Follow-up mediations were conducted using study measures available only at Time 2. For all study measures, Time 2 well-being independently predicted changes from no-to-yes responding by way of indirect effects through self-reported non-specific internalizing distress and arousal.

Conclusions: These findings highlight the confounding effects of fluctuations in current emotional distress on past stressor recall. There is a need for additional caution regarding the use of retrospective self-report of adverse life events in research and clinical practice and greater consideration of current psychological distress at the time of measurement completion.