Cardiovascular recovery from acute psychological stress: Testing the temporal stability of multiple models of recovery.

Objective: The variability in the operationalisation of cardiovascular recovery from psychological stress across studies is well-established in the field but little is done to address these issues. The aim of this study was to establish the temporal stability of multiple computations of cardiovascular recovery already documented in the literature. Secondly, the aim was to determine if temporal stability depended on how recovery was computed.

Methods: This study used previously collected data from the Pittsburgh Cold Study 3 (PCS3). One-hundred and eighty-nine participants (58.7 % male, two thirds Caucasian, age range; 18-55 years) completed two separate, identical, standardized stress-testing protocols (Mean = 48 days apart) and had their blood pressure and heart rate (HR) monitored throughout a baseline, stress task and recovery phase. Drawing on a multiverse framework, recovery was computed as 1) delta change score from baseline, 2) delta change score from task, 3) percent change baseline, 4) percent change task, 5) area under the curve with respect to ground (AUCg) and 6) area under the curve with respect to increase (AUCi).

Results: Delta change score from task, percent change task and AUCg demonstrated the strongest temporal stability for blood pressure and HR recovery between visits.

Conclusions: This study highlights the fragility of significant findings when multiple computations of cardiovascular recovery are considered within the multiverse framework, confirming that the status of results strongly depends on the computation of recovery that is chosen. We recommend employing multiverse analyses where feasible to confirm which computations are preferable. Where the multiverse is not feasible, we suggest using more sophisticated curve-fitting techniques like AUCg.